MyArxiv
Computer Vision and Pattern Recognition 96
☆ FaceXBench: Evaluating Multimodal LLMs on Face Understanding
Multimodal Large Language Models (MLLMs) demonstrate impressive problem-solving abilities across a wide range of tasks and domains. However, their capacity for face understanding has not been systematically studied. To address this gap, we introduce FaceXBench, a comprehensive benchmark designed to evaluate MLLMs on complex face understanding tasks. FaceXBench includes 5,000 multimodal multiple-choice questions derived from 25 public datasets and a newly created dataset, FaceXAPI. These questions cover 14 tasks across 6 broad categories, assessing MLLMs' face understanding abilities in bias and fairness, face authentication, recognition, analysis, localization and tool retrieval. Using FaceXBench, we conduct an extensive evaluation of 26 open-source MLLMs alongside 2 proprietary models, revealing the unique challenges in complex face understanding tasks. We analyze the models across three evaluation settings: zero-shot, in-context task description, and chain-of-thought prompting. Our detailed analysis reveals that current MLLMs, including advanced models like GPT-4o, and GeminiPro 1.5, show significant room for improvement. We believe FaceXBench will be a crucial resource for developing MLLMs equipped to perform sophisticated face understanding. Code: https://github.com/Kartik-3004/facexbench
comment: Project Page: https://kartik-3004.github.io/facexbench/
☆ Zero-Shot Monocular Scene Flow Estimation in the Wild
Large models have shown generalization across datasets for many low-level vision tasks, like depth estimation, but no such general models exist for scene flow. Even though scene flow has wide potential use, it is not used in practice because current predictive models do not generalize well. We identify three key challenges and propose solutions for each.First, we create a method that jointly estimates geometry and motion for accurate prediction. Second, we alleviate scene flow data scarcity with a data recipe that affords us 1M annotated training samples across diverse synthetic scenes. Third, we evaluate different parameterizations for scene flow prediction and adopt a natural and effective parameterization. Our resulting model outperforms existing methods as well as baselines built on large-scale models in terms of 3D end-point error, and shows zero-shot generalization to the casually captured videos from DAVIS and the robotic manipulation scenes from RoboTAP. Overall, our approach makes scene flow prediction more practical in-the-wild.
comment: Project Website: https://research.nvidia.com/labs/zero_msf
☆ 3rd Workshop on Maritime Computer Vision (MaCVi) 2025: Challenge Results
The 3rd Workshop on Maritime Computer Vision (MaCVi) 2025 addresses maritime computer vision for Unmanned Surface Vehicles (USV) and underwater. This report offers a comprehensive overview of the findings from the challenges. We provide both statistical and qualitative analyses, evaluating trends from over 700 submissions. All datasets, evaluation code, and the leaderboard are available to the public at https://macvi.org/workshop/macvi25.
comment: Part of the MaCVi 2025 workshop
☆ DiffStereo: High-Frequency Aware Diffusion Model for Stereo Image Restoration
Diffusion models (DMs) have achieved promising performance in image restoration but haven't been explored for stereo images. The application of DM in stereo image restoration is confronted with a series of challenges. The need to reconstruct two images exacerbates DM's computational cost. Additionally, existing latent DMs usually focus on semantic information and remove high-frequency details as redundancy during latent compression, which is precisely what matters for image restoration. To address the above problems, we propose a high-frequency aware diffusion model, DiffStereo for stereo image restoration as the first attempt at DM in this domain. Specifically, DiffStereo first learns latent high-frequency representations (LHFR) of HQ images. DM is then trained in the learned space to estimate LHFR for stereo images, which are fused into a transformer-based stereo image restoration network providing beneficial high-frequency information of corresponding HQ images. The resolution of LHFR is kept the same as input images, which preserves the inherent texture from distortion. And the compression in channels alleviates the computational burden of DM. Furthermore, we devise a position encoding scheme when integrating the LHFR into the restoration network, enabling distinctive guidance in different depths of the restoration network. Comprehensive experiments verify that by combining generative DM and transformer, DiffStereo achieves both higher reconstruction accuracy and better perceptual quality on stereo super-resolution, deblurring, and low-light enhancement compared with state-of-the-art methods.
comment: 9 pages, 6 figures
☆ New Fashion Products Performance Forecasting: A Survey on Evolutions, Models and Emerging Trends
The fast fashion industry's insatiable demand for new styles and rapid production cycles has led to a significant environmental burden. Overproduction, excessive waste, and harmful chemicals have contributed to the negative environmental impact of the industry. To mitigate these issues, a paradigm shift that prioritizes sustainability and efficiency is urgently needed. Integrating learning-based predictive analytics into the fashion industry represents a significant opportunity to address environmental challenges and drive sustainable practices. By forecasting fashion trends and optimizing production, brands can reduce their ecological footprint while remaining competitive in a rapidly changing market. However, one of the key challenges in forecasting fashion sales is the dynamic nature of consumer preferences. Fashion is acyclical, with trends constantly evolving and resurfacing. In addition, cultural changes and unexpected events can disrupt established patterns. This problem is also known as New Fashion Products Performance Forecasting (NFPPF), and it has recently gained more and more interest in the global research landscape. Given its multidisciplinary nature, the field of NFPPF has been approached from many different angles. This comprehensive survey wishes to provide an up-to-date overview that focuses on learning-based NFPPF strategies. The survey is based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodological flow, allowing for a systematic and complete literature review. In particular, we propose the first taxonomy that covers the learning panorama for NFPPF, examining in detail the different methodologies used to increase the amount of multimodal information, as well as the state-of-the-art available datasets. Finally, we discuss the challenges and future directions.
comment: Accepted at the Springer Nature Computer Science journal
☆ HiMix: Reducing Computational Complexity in Large Vision-Language Models
Benefiting from recent advancements in large language models and modality alignment techniques, existing Large Vision-Language Models(LVLMs) have achieved prominent performance across a wide range of scenarios. However, the excessive computational complexity limits the widespread use of these models in practical applications. We argue that one main bottleneck in computational complexity is caused by the involvement of redundant vision sequences in model computation. This is inspired by a reassessment of the efficiency of vision and language information transmission in the language decoder of LVLMs. Then, we propose a novel hierarchical vision-language interaction mechanism called Hierarchical Vision injection for Mixture Attention (HiMix). In HiMix, only the language sequence undergoes full forward propagation, while the vision sequence interacts with the language at specific stages within each language decoder layer. It is striking that our approach significantly reduces computational complexity with minimal performance loss. Specifically, HiMix achieves a 10x reduction in the computational cost of the language decoder across multiple LVLM models while maintaining comparable performance. This highlights the advantages of our method, and we hope our research brings new perspectives to the field of vision-language understanding. Project Page: https://xuange923.github.io/HiMix
☆ GSTAR: Gaussian Surface Tracking and Reconstruction
3D Gaussian Splatting techniques have enabled efficient photo-realistic rendering of static scenes. Recent works have extended these approaches to support surface reconstruction and tracking. However, tracking dynamic surfaces with 3D Gaussians remains challenging due to complex topology changes, such as surfaces appearing, disappearing, or splitting. To address these challenges, we propose GSTAR, a novel method that achieves photo-realistic rendering, accurate surface reconstruction, and reliable 3D tracking for general dynamic scenes with changing topology. Given multi-view captures as input, GSTAR binds Gaussians to mesh faces to represent dynamic objects. For surfaces with consistent topology, GSTAR maintains the mesh topology and tracks the meshes using Gaussians. In regions where topology changes, GSTAR adaptively unbinds Gaussians from the mesh, enabling accurate registration and the generation of new surfaces based on these optimized Gaussians. Additionally, we introduce a surface-based scene flow method that provides robust initialization for tracking between frames. Experiments demonstrate that our method effectively tracks and reconstructs dynamic surfaces, enabling a range of applications. Our project page with the code release is available at https://chengwei-zheng.github.io/GSTAR/.
☆ MutualForce: Mutual-Aware Enhancement for 4D Radar-LiDAR 3D Object Detection ICASSP 2025
Radar and LiDAR have been widely used in autonomous driving as LiDAR provides rich structure information, and radar demonstrates high robustness under adverse weather. Recent studies highlight the effectiveness of fusing radar and LiDAR point clouds. However, challenges remain due to the modality misalignment and information loss during feature extractions. To address these issues, we propose a 4D radar-LiDAR framework to mutually enhance their representations. Initially, the indicative features from radar are utilized to guide both radar and LiDAR geometric feature learning. Subsequently, to mitigate their sparsity gap, the shape information from LiDAR is used to enrich radar BEV features. Extensive experiments on the View-of-Delft (VoD) dataset demonstrate our approach's superiority over existing methods, achieving the highest mAP of 71.76% across the entire area and 86.36\% within the driving corridor. Especially for cars, we improve the AP by 4.17% and 4.20% due to the strong indicative features and symmetric shapes.
comment: Accepted by ICASSP 2025
☆ Robust Egoistic Rigid Body Localization
We consider a robust and self-reliant (or "egoistic") variation of the rigid body localization (RBL) problem, in which a primary rigid body seeks to estimate the pose (i.e., location and orientation) of another rigid body (or "target"), relative to its own, without the assistance of external infrastructure, without prior knowledge of the shape of the target, and taking into account the possibility that the available observations are incomplete. Three complementary contributions are then offered for such a scenario. The first is a method to estimate the translation vector between the center point of both rigid bodies, which unlike existing techniques does not require that both objects have the same shape or even the same number of landmark points. This technique is shown to significantly outperform the state-of-the-art (SotA) under complete information, but to be sensitive to data erasures, even when enhanced by matrix completion methods. The second contribution, designed to offer improved performance in the presence of incomplete information, offers a robust alternative to the latter, at the expense of a slight relative loss under complete information. Finally, the third contribution is a scheme for the estimation of the rotation matrix describing the relative orientation of the target rigid body with respect to the primary. Comparisons of the proposed schemes and SotA techniques demonstrate the advantage of the contributed methods in terms of root mean square error (RMSE) performance under fully complete information and incomplete conditions.
☆ Disharmony: Forensics using Reverse Lighting Harmonization
Content generation and manipulation approaches based on deep learning methods have seen significant advancements, leading to an increased need for techniques to detect whether an image has been generated or edited. Another area of research focuses on the insertion and harmonization of objects within images. In this study, we explore the potential of using harmonization data in conjunction with a segmentation model to enhance the detection of edited image regions. These edits can be either manually crafted or generated using deep learning methods. Our findings demonstrate that this approach can effectively identify such edits. Existing forensic models often overlook the detection of harmonized objects in relation to the background, but our proposed Disharmony Network addresses this gap. By utilizing an aggregated dataset of harmonization techniques, our model outperforms existing forensic networks in identifying harmonized objects integrated into their backgrounds, and shows potential for detecting various forms of edits, including virtual try-on tasks.
☆ Hypercone Assisted Contour Generation for Out-of-Distribution Detection
Recent advances in the field of out-of-distribution (OOD) detection have placed great emphasis on learning better representations suited to this task. While there are distance-based approaches, distributional awareness has seldom been exploited for better performance. We present HAC$_k$-OOD, a novel OOD detection method that makes no distributional assumption about the data, but automatically adapts to its distribution. Specifically, HAC$_k$-OOD constructs a set of hypercones by maximizing the angular distance to neighbors in a given data-point's vicinity to approximate the contour within which in-distribution (ID) data-points lie. Experimental results show state-of-the-art FPR@95 and AUROC performance on Near-OOD detection and on Far-OOD detection on the challenging CIFAR-100 benchmark without explicitly training for OOD performance.
☆ Adaptive Clustering for Efficient Phenotype Segmentation of UAV Hyperspectral Data WACV 2025
Unmanned Aerial Vehicles (UAVs) combined with Hyperspectral imaging (HSI) offer potential for environmental and agricultural applications by capturing detailed spectral information that enables the prediction of invisible features like biochemical leaf properties. However, the data-intensive nature of HSI poses challenges for remote devices, which have limited computational resources and storage. This paper introduces an Online Hyperspectral Simple Linear Iterative Clustering algorithm (OHSLIC) framework for real-time tree phenotype segmentation. OHSLIC reduces inherent noise and computational demands through adaptive incremental clustering and a lightweight neural network, which phenotypes trees using leaf contents such as chlorophyll, carotenoids, and anthocyanins. A hyperspectral dataset is created using a custom simulator that incorporates realistic leaf parameters, and light interactions. Results demonstrate that OHSLIC achieves superior regression accuracy and segmentation performance compared to pixel- or window-based methods while significantly reducing inference time. The method`s adaptive clustering enables dynamic trade-offs between computational efficiency and accuracy, paving the way for scalable edge-device deployment in HSI applications.
comment: accepted WACV 2025 GeoCV workshop
☆ CSHNet: A Novel Information Asymmetric Image Translation Method
Despite advancements in cross-domain image translation, challenges persist in asymmetric tasks such as SAR-to-Optical and Sketch-to-Instance conversions, which involve transforming data from a less detailed domain into one with richer content. Traditional CNN-based methods are effective at capturing fine details but struggle with global structure, leading to unwanted merging of image regions. To address this, we propose the CNN-Swin Hybrid Network (CSHNet), which combines two key modules: Swin Embedded CNN (SEC) and CNN Embedded Swin (CES), forming the SEC-CES-Bottleneck (SCB). SEC leverages CNN's detailed feature extraction while integrating the Swin Transformer's structural bias. CES, in turn, preserves the Swin Transformer's global integrity, compensating for CNN's lack of focus on structure. Additionally, CSHNet includes two components designed to enhance cross-domain information retention: the Interactive Guided Connection (IGC), which enables dynamic information exchange between SEC and CES, and Adaptive Edge Perception Loss (AEPL), which maintains structural boundaries during translation. Experimental results show that CSHNet outperforms existing methods in both visual quality and performance metrics across scene-level and instance-level datasets. Our code is available at: https://github.com/XduShi/CSHNet.
☆ Structure-guided Deep Multi-View Clustering
Deep multi-view clustering seeks to utilize the abundant information from multiple views to improve clustering performance. However, most of the existing clustering methods often neglect to fully mine multi-view structural information and fail to explore the distribution of multi-view data, limiting clustering performance. To address these limitations, we propose a structure-guided deep multi-view clustering model. Specifically, we introduce a positive sample selection strategy based on neighborhood relationships, coupled with a corresponding loss function. This strategy constructs multi-view nearest neighbor graphs to dynamically redefine positive sample pairs, enabling the mining of local structural information within multi-view data and enhancing the reliability of positive sample selection. Additionally, we introduce a Gaussian distribution model to uncover latent structural information and introduce a loss function to reduce discrepancies between view embeddings. These two strategies explore multi-view structural information and data distribution from different perspectives, enhancing consistency across views and increasing intra-cluster compactness. Experimental evaluations demonstrate the efficacy of our method, showing significant improvements in clustering performance on multiple benchmark datasets compared to state-of-the-art multi-view clustering approaches.
☆ A Vision-Language Framework for Multispectral Scene Representation Using Language-Grounded Features
Scene understanding in remote sensing often faces challenges in generating accurate representations for complex environments such as various land use areas or coastal regions, which may also include snow, clouds, or haze. To address this, we present a vision-language framework named Spectral LLaVA, which integrates multispectral data with vision-language alignment techniques to enhance scene representation and description. Using the BigEarthNet v2 dataset from Sentinel-2, we establish a baseline with RGB-based scene descriptions and further demonstrate substantial improvements through the incorporation of multispectral information. Our framework optimizes a lightweight linear projection layer for alignment while keeping the vision backbone of SpectralGPT frozen. Our experiments encompass scene classification using linear probing and language modeling for jointly performing scene classification and description generation. Our results highlight Spectral LLaVA's ability to produce detailed and accurate descriptions, particularly for scenarios where RGB data alone proves inadequate, while also enhancing classification performance by refining SpectralGPT features into semantically meaningful representations.
☆ ACE: Anatomically Consistent Embeddings in Composition and Decomposition WACV 2025
Medical images acquired from standardized protocols show consistent macroscopic or microscopic anatomical structures, and these structures consist of composable/decomposable organs and tissues, but existing self-supervised learning (SSL) methods do not appreciate such composable/decomposable structure attributes inherent to medical images. To overcome this limitation, this paper introduces a novel SSL approach called ACE to learn anatomically consistent embedding via composition and decomposition with two key branches: (1) global consistency, capturing discriminative macro-structures via extracting global features; (2) local consistency, learning fine-grained anatomical details from composable/decomposable patch features via corresponding matrix matching. Experimental results across 6 datasets 2 backbones, evaluated in few-shot learning, fine-tuning, and property analysis, show ACE's superior robustness, transferability, and clinical potential. The innovations of our ACE lie in grid-wise image cropping, leveraging the intrinsic properties of compositionality and decompositionality of medical images, bridging the semantic gap from high-level pathologies to low-level tissue anomalies, and providing a new SSL method for medical imaging.
comment: Accepted by WACV 2025
☆ Spatio-temporal Graph Learning on Adaptive Mined Key Frames for High-performance Multi-Object Tracking
In the realm of multi-object tracking, the challenge of accurately capturing the spatial and temporal relationships between objects in video sequences remains a significant hurdle. This is further complicated by frequent occurrences of mutual occlusions among objects, which can lead to tracking errors and reduced performance in existing methods. Motivated by these challenges, we propose a novel adaptive key frame mining strategy that addresses the limitations of current tracking approaches. Specifically, we introduce a Key Frame Extraction (KFE) module that leverages reinforcement learning to adaptively segment videos, thereby guiding the tracker to exploit the intrinsic logic of the video content. This approach allows us to capture structured spatial relationships between different objects as well as the temporal relationships of objects across frames. To tackle the issue of object occlusions, we have developed an Intra-Frame Feature Fusion (IFF) module. Unlike traditional graph-based methods that primarily focus on inter-frame feature fusion, our IFF module uses a Graph Convolutional Network (GCN) to facilitate information exchange between the target and surrounding objects within a frame. This innovation significantly enhances target distinguishability and mitigates tracking loss and appearance similarity due to occlusions. By combining the strengths of both long and short trajectories and considering the spatial relationships between objects, our proposed tracker achieves impressive results on the MOT17 dataset, i.e., 68.6 HOTA, 81.0 IDF1, 66.6 AssA, and 893 IDS, proving its effectiveness and accuracy.
☆ FECT: Classification of Breast Cancer Pathological Images Based on Fusion Features
Breast cancer is one of the most common cancers among women globally, with early diagnosis and precise classification being crucial. With the advancement of deep learning and computer vision, the automatic classification of breast tissue pathological images has emerged as a research focus. Existing methods typically rely on singular cell or tissue features and lack design considerations for morphological characteristics of challenging-to-classify categories, resulting in suboptimal classification performance. To address these problems, we proposes a novel breast cancer tissue classification model that Fused features of Edges, Cells, and Tissues (FECT), employing the ResMTUNet and an attention-based aggregator to extract and aggregate these features. Extensive testing on the BRACS dataset demonstrates that our model surpasses current advanced methods in terms of classification accuracy and F1 scores. Moreover, due to its feature fusion that aligns with the diagnostic approach of pathologists, our model exhibits interpretability and holds promise for significant roles in future clinical applications.
☆ DiffVSR: Enhancing Real-World Video Super-Resolution with Diffusion Models for Advanced Visual Quality and Temporal Consistency
Diffusion models have demonstrated exceptional capabilities in image generation and restoration, yet their application to video super-resolution faces significant challenges in maintaining both high fidelity and temporal consistency. We present DiffVSR, a diffusion-based framework for real-world video super-resolution that effectively addresses these challenges through key innovations. For intra-sequence coherence, we develop a multi-scale temporal attention module and temporal-enhanced VAE decoder that capture fine-grained motion details. To ensure inter-sequence stability, we introduce a noise rescheduling mechanism with an interweaved latent transition approach, which enhances temporal consistency without additional training overhead. We propose a progressive learning strategy that transitions from simple to complex degradations, enabling robust optimization despite limited high-quality video data. Extensive experiments demonstrate that DiffVSR delivers superior results in both visual quality and temporal consistency, setting a new performance standard in real-world video super-resolution.
comment: Project page: \url{https://xh9998.github.io/DiffVSR-project/}
☆ Universal Actions for Enhanced Embodied Foundation Models
Training on diverse, internet-scale data is a key factor in the success of recent large foundation models. Yet, using the same recipe for building embodied agents has faced noticeable difficulties. Despite the availability of many crowd-sourced embodied datasets, their action spaces often exhibit significant heterogeneity due to distinct physical embodiment and control interfaces for different robots, causing substantial challenges in developing embodied foundation models using cross-domain data. In this paper, we introduce UniAct, a new embodied foundation modeling framework operating in a tokenized Universal Action Space. Our learned universal actions capture the generic atomic behaviors across diverse robots by exploiting their shared structural features, and enable enhanced cross-domain data utilization and cross-embodiment generalizations by eliminating the notorious heterogeneity. The universal actions can be efficiently translated back to heterogeneous actionable commands by simply adding embodiment-specific details, from which fast adaptation to new robots becomes simple and straightforward. Our 0.5B instantiation of UniAct outperforms 14X larger SOTA embodied foundation models in extensive evaluations on various real-world and simulation robots, showcasing exceptional cross-embodiment control and adaptation capability, highlighting the crucial benefit of adopting universal actions. Project page: https://github.com/2toinf/UniAct
comment: Preprint
☆ landmarker: a Toolkit for Anatomical Landmark Localization in 2D/3D Images
Anatomical landmark localization in 2D/3D images is a critical task in medical imaging. Although many general-purpose tools exist for landmark localization in classical computer vision tasks, such as pose estimation, they lack the specialized features and modularity necessary for anatomical landmark localization applications in the medical domain. Therefore, we introduce landmarker, a Python package built on PyTorch. The package provides a comprehensive, flexible toolkit for developing and evaluating landmark localization algorithms, supporting a range of methodologies, including static and adaptive heatmap regression. landmarker enhances the accuracy of landmark identification, streamlines research and development processes, and supports various image formats and preprocessing pipelines. Its modular design allows users to customize and extend the toolkit for specific datasets and applications, accelerating innovation in medical imaging. landmarker addresses a critical need for precision and customization in landmark localization tasks not adequately met by existing general-purpose pose estimation tools.
comment: 11 pages, 4 figures
☆ Classifier Ensemble for Efficient Uncertainty Calibration of Deep Neural Networks for Image Classification
This paper investigates novel classifier ensemble techniques for uncertainty calibration applied to various deep neural networks for image classification. We evaluate both accuracy and calibration metrics, focusing on Expected Calibration Error (ECE) and Maximum Calibration Error (MCE). Our work compares different methods for building simple yet efficient classifier ensembles, including majority voting and several metamodel-based approaches. Our evaluation reveals that while state-of-the-art deep neural networks for image classification achieve high accuracy on standard datasets, they frequently suffer from significant calibration errors. Basic ensemble techniques like majority voting provide modest improvements, while metamodel-based ensembles consistently reduce ECE and MCE across all architectures. Notably, the largest of our compared metamodels demonstrate the most substantial calibration improvements, with minimal impact on accuracy. Moreover, classifier ensembles with metamodels outperform traditional model ensembles in calibration performance, while requiring significantly fewer parameters. In comparison to traditional post-hoc calibration methods, our approach removes the need for a separate calibration dataset. These findings underscore the potential of our proposed metamodel-based classifier ensembles as an efficient and effective approach to improving model calibration, thereby contributing to more reliable deep learning systems.
comment: This paper has been accepted at International Conference on Computer Vision Theory and Applications (VISAPP), 2025
☆ Leveraging Confident Image Regions for Source-Free Domain-Adaptive Object Detection
Source-free domain-adaptive object detection is an interesting but scarcely addressed topic. It aims at adapting a source-pretrained detector to a distinct target domain without resorting to source data during adaptation. So far, there is no data augmentation scheme tailored to source-free domain-adaptive object detection. To this end, this paper presents a novel data augmentation approach that cuts out target image regions where the detector is confident, augments them along with their respective pseudo-labels, and joins them into a challenging target image to adapt the detector. As the source data is out of reach during adaptation, we implement our approach within a teacher-student learning paradigm to ensure that the model does not collapse during the adaptation procedure. We evaluated our approach on three adaptation benchmarks of traffic scenes, scoring new state-of-the-art on two of them.
☆ Few-shot Structure-Informed Machinery Part Segmentation with Foundation Models and Graph Neural Networks WACV
This paper proposes a novel approach to few-shot semantic segmentation for machinery with multiple parts that exhibit spatial and hierarchical relationships. Our method integrates the foundation models CLIPSeg and Segment Anything Model (SAM) with the interest point detector SuperPoint and a graph convolutional network (GCN) to accurately segment machinery parts. By providing 1 to 25 annotated samples, our model, evaluated on a purely synthetic dataset depicting a truck-mounted loading crane, achieves effective segmentation across various levels of detail. Training times are kept under five minutes on consumer GPUs. The model demonstrates robust generalization to real data, achieving a qualitative synthetic-to-real generalization with a $J\&F$ score of 92.2 on real data using 10 synthetic support samples. When benchmarked on the DAVIS 2017 dataset, it achieves a $J\&F$ score of 71.5 in semi-supervised video segmentation with three support samples. This method's fast training times and effective generalization to real data make it a valuable tool for autonomous systems interacting with machinery and infrastructure, and illustrate the potential of combined and orchestrated foundation models for few-shot segmentation tasks.
comment: Accepted at Winter Conference on Applications of Computer Vision (WACV) 2025. Code and available at https://github.com/AIT-Assistive-Autonomous-Systems/Hopomop
☆ Robust Change Captioning in Remote Sensing: SECOND-CC Dataset and MModalCC Framework
Remote sensing change captioning (RSICC) aims to describe changes between bitemporal images in natural language. Existing methods often fail under challenges like illumination differences, viewpoint changes, blur effects, leading to inaccuracies, especially in no-change regions. Moreover, the images acquired at different spatial resolutions and have registration errors tend to affect the captions. To address these issues, we introduce SECOND-CC, a novel RSICC dataset featuring high-resolution RGB image pairs, semantic segmentation maps, and diverse real-world scenarios. SECOND-CC which contains 6,041 pairs of bitemporal RS images and 30,205 sentences describing the differences between images. Additionally, we propose MModalCC, a multimodal framework that integrates semantic and visual data using advanced attention mechanisms, including Cross-Modal Cross Attention (CMCA) and Multimodal Gated Cross Attention (MGCA). Detailed ablation studies and attention visualizations further demonstrate its effectiveness and ability to address RSICC challenges. Comprehensive experiments show that MModalCC outperforms state-of-the-art RSICC methods, including RSICCformer, Chg2Cap, and PSNet with +4.6% improvement on BLEU4 score and +9.6% improvement on CIDEr score. We will make our dataset and codebase publicly available to facilitate future research at https://github.com/ChangeCapsInRS/SecondCC
comment: This work has been submitted to the IEEE Transactions on Geoscience and Remote Sensing journal for possible publication
☆ SpatialCoT: Advancing Spatial Reasoning through Coordinate Alignment and Chain-of-Thought for Embodied Task Planning
Spatial reasoning is an essential problem in embodied AI research. Efforts to enhance spatial reasoning abilities through supplementary spatial data and fine-tuning have proven limited and ineffective when addressing complex embodied tasks, largely due to their dependence on language-based outputs. While some approaches have introduced a point-based action space to mitigate this issue, they fall short in managing more intricate tasks within complex environments. This deficiency arises from their failure to fully exploit the inherent thinking and reasoning capabilities that are fundamental strengths of Vision-Language Models (VLMs). To address these limitations, we propose a novel approach named SpatialCoT, specifically designed to bolster the spatial reasoning capabilities of VLMs. Our approach comprises two stages: spatial coordinate bi-directional alignment, which aligns vision-language inputs with spatial coordinates, and chain-of-thought spatial grounding, which harnesses the reasoning capabilities of language models for advanced spatial reasoning. We evaluate SpatialCoT on challenging navigation and manipulation tasks, both in simulation and real-world settings. Experimental results demonstrate that our method significantly outperforms previous state-of-the-art approaches in both tasks.
comment: 13 pages, 6 figures
☆ CLIP-PCQA: Exploring Subjective-Aligned Vision-Language Modeling for Point Cloud Quality Assessment
In recent years, No-Reference Point Cloud Quality Assessment (NR-PCQA) research has achieved significant progress. However, existing methods mostly seek a direct mapping function from visual data to the Mean Opinion Score (MOS), which is contradictory to the mechanism of practical subjective evaluation. To address this, we propose a novel language-driven PCQA method named CLIP-PCQA. Considering that human beings prefer to describe visual quality using discrete quality descriptions (e.g., "excellent" and "poor") rather than specific scores, we adopt a retrieval-based mapping strategy to simulate the process of subjective assessment. More specifically, based on the philosophy of CLIP, we calculate the cosine similarity between the visual features and multiple textual features corresponding to different quality descriptions, in which process an effective contrastive loss and learnable prompts are introduced to enhance the feature extraction. Meanwhile, given the personal limitations and bias in subjective experiments, we further covert the feature similarities into probabilities and consider the Opinion Score Distribution (OSD) rather than a single MOS as the final target. Experimental results show that our CLIP-PCQA outperforms other State-Of-The-Art (SOTA) approaches.
☆ FiLo++: Zero-/Few-Shot Anomaly Detection by Fused Fine-Grained Descriptions and Deformable Localization
Anomaly detection methods typically require extensive normal samples from the target class for training, limiting their applicability in scenarios that require rapid adaptation, such as cold start. Zero-shot and few-shot anomaly detection do not require labeled samples from the target class in advance, making them a promising research direction. Existing zero-shot and few-shot approaches often leverage powerful multimodal models to detect and localize anomalies by comparing image-text similarity. However, their handcrafted generic descriptions fail to capture the diverse range of anomalies that may emerge in different objects, and simple patch-level image-text matching often struggles to localize anomalous regions of varying shapes and sizes. To address these issues, this paper proposes the FiLo++ method, which consists of two key components. The first component, Fused Fine-Grained Descriptions (FusDes), utilizes large language models to generate anomaly descriptions for each object category, combines both fixed and learnable prompt templates and applies a runtime prompt filtering method, producing more accurate and task-specific textual descriptions. The second component, Deformable Localization (DefLoc), integrates the vision foundation model Grounding DINO with position-enhanced text descriptions and a Multi-scale Deformable Cross-modal Interaction (MDCI) module, enabling accurate localization of anomalies with various shapes and sizes. In addition, we design a position-enhanced patch matching approach to improve few-shot anomaly detection performance. Experiments on multiple datasets demonstrate that FiLo++ achieves significant performance improvements compared with existing methods. Code will be available at https://github.com/CASIA-IVA-Lab/FiLo.
☆ One-D-Piece: Image Tokenizer Meets Quality-Controllable Compression
Current image tokenization methods require a large number of tokens to capture the information contained within images. Although the amount of information varies across images, most image tokenizers only support fixed-length tokenization, leading to inefficiency in token allocation. In this study, we introduce One-D-Piece, a discrete image tokenizer designed for variable-length tokenization, achieving quality-controllable mechanism. To enable variable compression rate, we introduce a simple but effective regularization mechanism named "Tail Token Drop" into discrete one-dimensional image tokenizers. This method encourages critical information to concentrate at the head of the token sequence, enabling support of variadic tokenization, while preserving state-of-the-art reconstruction quality. We evaluate our tokenizer across multiple reconstruction quality metrics and find that it delivers significantly better perceptual quality than existing quality-controllable compression methods, including JPEG and WebP, at smaller byte sizes. Furthermore, we assess our tokenizer on various downstream computer vision tasks, including image classification, object detection, semantic segmentation, and depth estimation, confirming its adaptability to numerous applications compared to other variable-rate methods. Our approach demonstrates the versatility of variable-length discrete image tokenization, establishing a new paradigm in both compression efficiency and reconstruction performance. Finally, we validate the effectiveness of tail token drop via detailed analysis of tokenizers.
comment: Our Project Page: https://turingmotors.github.io/one-d-piece-tokenizer
☆ LWGANet: A Lightweight Group Attention Backbone for Remote Sensing Visual Tasks
Remote sensing (RS) visual tasks have gained significant academic and practical importance. However, they encounter numerous challenges that hinder effective feature extraction, including the detection and recognition of multiple objects exhibiting substantial variations in scale within a single image. While prior dual-branch or multi-branch architectural strategies have been effective in managing these object variances, they have concurrently resulted in considerable increases in computational demands and parameter counts. Consequently, these architectures are rendered less viable for deployment on resource-constrained devices. Contemporary lightweight backbone networks, designed primarily for natural images, frequently encounter difficulties in effectively extracting features from multi-scale objects, which compromises their efficacy in RS visual tasks. This article introduces LWGANet, a specialized lightweight backbone network tailored for RS visual tasks, incorporating a novel lightweight group attention (LWGA) module designed to address these specific challenges. LWGA module, tailored for RS imagery, adeptly harnesses redundant features to extract a wide range of spatial information, from local to global scales, without introducing additional complexity or computational overhead. This facilitates precise feature extraction across multiple scales within an efficient framework.LWGANet was rigorously evaluated across twelve datasets, which span four crucial RS visual tasks: scene classification, oriented object detection, semantic segmentation, and change detection. The results confirm LWGANet's widespread applicability and its ability to maintain an optimal balance between high performance and low complexity, achieving SOTA results across diverse datasets. LWGANet emerged as a novel solution for resource-limited scenarios requiring robust RS image processing capabilities.
comment: 12 pages, 8 figures, Remote sensing
☆ X-Dyna: Expressive Dynamic Human Image Animation
We introduce X-Dyna, a novel zero-shot, diffusion-based pipeline for animating a single human image using facial expressions and body movements derived from a driving video, that generates realistic, context-aware dynamics for both the subject and the surrounding environment. Building on prior approaches centered on human pose control, X-Dyna addresses key shortcomings causing the loss of dynamic details, enhancing the lifelike qualities of human video animations. At the core of our approach is the Dynamics-Adapter, a lightweight module that effectively integrates reference appearance context into the spatial attentions of the diffusion backbone while preserving the capacity of motion modules in synthesizing fluid and intricate dynamic details. Beyond body pose control, we connect a local control module with our model to capture identity-disentangled facial expressions, facilitating accurate expression transfer for enhanced realism in animated scenes. Together, these components form a unified framework capable of learning physical human motion and natural scene dynamics from a diverse blend of human and scene videos. Comprehensive qualitative and quantitative evaluations demonstrate that X-Dyna outperforms state-of-the-art methods, creating highly lifelike and expressive animations. The code is available at https://github.com/bytedance/X-Dyna.
comment: Project page:https://x-dyna.github.io/xdyna.github.io/ Code:https://github.com/bytedance/X-Dyna
☆ Textoon: Generating Vivid 2D Cartoon Characters from Text Descriptions
The 2D cartoon style is a prominent art form in digital character creation, particularly popular among younger audiences. While advancements in digital human technology have spurred extensive research into photorealistic digital humans and 3D characters, interactive 2D cartoon characters have received comparatively less attention. Unlike 3D counterparts, which require sophisticated construction and resource-intensive rendering, Live2D, a widely-used format for 2D cartoon characters, offers a more efficient alternative, which allows to animate 2D characters in a manner that simulates 3D movement without the necessity of building a complete 3D model. Furthermore, Live2D employs lightweight HTML5 (H5) rendering, improving both accessibility and efficiency. In this technical report, we introduce Textoon, an innovative method for generating diverse 2D cartoon characters in the Live2D format based on text descriptions. The Textoon leverages cutting-edge language and vision models to comprehend textual intentions and generate 2D appearance, capable of creating a wide variety of stunning and interactive 2D characters within one minute. The project homepage is https://human3daigc.github.io/Textoon_webpage/.
☆ DiffuEraser: A Diffusion Model for Video Inpainting
Recent video inpainting algorithms integrate flow-based pixel propagation with transformer-based generation to leverage optical flow for restoring textures and objects using information from neighboring frames, while completing masked regions through visual Transformers. However, these approaches often encounter blurring and temporal inconsistencies when dealing with large masks, highlighting the need for models with enhanced generative capabilities. Recently, diffusion models have emerged as a prominent technique in image and video generation due to their impressive performance. In this paper, we introduce DiffuEraser, a video inpainting model based on stable diffusion, designed to fill masked regions with greater details and more coherent structures. We incorporate prior information to provide initialization and weak conditioning,which helps mitigate noisy artifacts and suppress hallucinations. Additionally, to improve temporal consistency during long-sequence inference, we expand the temporal receptive fields of both the prior model and DiffuEraser, and further enhance consistency by leveraging the temporal smoothing property of Video Diffusion Models. Experimental results demonstrate that our proposed method outperforms state-of-the-art techniques in both content completeness and temporal consistency while maintaining acceptable efficiency.
comment: 11pages, 13figures
☆ Mitigating Hallucinations on Object Attributes using Multiview Images and Negative Instructions ICASSP 2025
Current popular Large Vision-Language Models (LVLMs) are suffering from Hallucinations on Object Attributes (HoOA), leading to incorrect determination of fine-grained attributes in the input images. Leveraging significant advancements in 3D generation from a single image, this paper proposes a novel method to mitigate HoOA in LVLMs. This method utilizes multiview images sampled from generated 3D representations as visual prompts for LVLMs, thereby providing more visual information from other viewpoints. Furthermore, we observe the input order of multiple multiview images significantly affects the performance of LVLMs. Consequently, we have devised Multiview Image Augmented VLM (MIAVLM), incorporating a Multiview Attributes Perceiver (MAP) submodule capable of simultaneously eliminating the influence of input image order and aligning visual information from multiview images with Large Language Models (LLMs). Besides, we designed and employed negative instructions to mitigate LVLMs' bias towards ``Yes" responses. Comprehensive experiments demonstrate the effectiveness of our method.
comment: 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025)
☆ Deep Learning for Early Alzheimer Disease Detection with MRI Scans
Alzheimer's Disease is a neurodegenerative condition characterized by dementia and impairment in neurological function. The study primarily focuses on the individuals above age 40, affecting their memory, behavior, and cognitive processes of the brain. Alzheimer's disease requires diagnosis by a detailed assessment of MRI scans and neuropsychological tests of the patients. This project compares existing deep learning models in the pursuit of enhancing the accuracy and efficiency of AD diagnosis, specifically focusing on the Convolutional Neural Network, Bayesian Convolutional Neural Network, and the U-net model with the Open Access Series of Imaging Studies brain MRI dataset. Besides, to ensure robustness and reliability in the model evaluations, we address the challenge of imbalance in data. We then perform rigorous evaluation to determine strengths and weaknesses for each model by considering sensitivity, specificity, and computational efficiency. This comparative analysis would shed light on the future role of AI in revolutionizing AD diagnostics but also paved ways for future innovation in medical imaging and the management of neurodegenerative diseases.
☆ Multi-Modal Attention Networks for Enhanced Segmentation and Depth Estimation of Subsurface Defects in Pulse Thermography
AI-driven pulse thermography (PT) has become a crucial tool in non-destructive testing (NDT), enabling automatic detection of hidden anomalies in various industrial components. Current state-of-the-art techniques feed segmentation and depth estimation networks compressed PT sequences using either Principal Component Analysis (PCA) or Thermographic Signal Reconstruction (TSR). However, treating these two modalities independently constrains the performance of PT inspection models as these representations possess complementary semantic features. To address this limitation, this work proposes PT-Fusion, a multi-modal attention-based fusion network that fuses both PCA and TSR modalities for defect segmentation and depth estimation of subsurface defects in PT setups. PT-Fusion introduces novel feature fusion modules, Encoder Attention Fusion Gate (EAFG) and Attention Enhanced Decoding Block (AEDB), to fuse PCA and TSR features for enhanced segmentation and depth estimation of subsurface defects. In addition, a novel data augmentation technique is proposed based on random data sampling from thermographic sequences to alleviate the scarcity of PT datasets. The proposed method is benchmarked against state-of-the-art PT inspection models, including U-Net, attention U-Net, and 3D-CNN on the Universit\'e Laval IRT-PVC dataset. The results demonstrate that PT-Fusion outperforms the aforementioned models in defect segmentation and depth estimation accuracies with a margin of 10%.
comment: Pulse thermography, infrared thermography, defect segmentation, multi-modal networks, attention mechanism
☆ RichSpace: Enriching Text-to-Video Prompt Space via Text Embedding Interpolation
Text-to-video generation models have made impressive progress, but they still struggle with generating videos with complex features. This limitation often arises from the inability of the text encoder to produce accurate embeddings, which hinders the video generation model. In this work, we propose a novel approach to overcome this challenge by selecting the optimal text embedding through interpolation in the embedding space. We demonstrate that this method enables the video generation model to produce the desired videos. Additionally, we introduce a simple algorithm using perpendicular foot embeddings and cosine similarity to identify the optimal interpolation embedding. Our findings highlight the importance of accurate text embeddings and offer a pathway for improving text-to-video generation performance.
☆ Aneumo: A Large-Scale Comprehensive Synthetic Dataset of Aneurysm Hemodynamics
Intracranial aneurysm (IA) is a common cerebrovascular disease that is usually asymptomatic but may cause severe subarachnoid hemorrhage (SAH) if ruptured. Although clinical practice is usually based on individual factors and morphological features of the aneurysm, its pathophysiology and hemodynamic mechanisms remain controversial. To address the limitations of current research, this study constructed a comprehensive hemodynamic dataset of intracranial aneurysms. The dataset is based on 466 real aneurysm models, and 10,000 synthetic models were generated by resection and deformation operations, including 466 aneurysm-free models and 9,534 deformed aneurysm models. The dataset also provides medical image-like segmentation mask files to support insightful analysis. In addition, the dataset contains hemodynamic data measured at eight steady-state flow rates (0.001 to 0.004 kg/s), including critical parameters such as flow velocity, pressure, and wall shear stress, providing a valuable resource for investigating aneurysm pathogenesis and clinical prediction. This dataset will help advance the understanding of the pathologic features and hemodynamic mechanisms of intracranial aneurysms and support in-depth research in related fields. Dataset hosted at https://github.com/Xigui-Li/Aneumo.
☆ GaussianAvatar-Editor: Photorealistic Animatable Gaussian Head Avatar Editor 3DV 2025
We introduce GaussianAvatar-Editor, an innovative framework for text-driven editing of animatable Gaussian head avatars that can be fully controlled in expression, pose, and viewpoint. Unlike static 3D Gaussian editing, editing animatable 4D Gaussian avatars presents challenges related to motion occlusion and spatial-temporal inconsistency. To address these issues, we propose the Weighted Alpha Blending Equation (WABE). This function enhances the blending weight of visible Gaussians while suppressing the influence on non-visible Gaussians, effectively handling motion occlusion during editing. Furthermore, to improve editing quality and ensure 4D consistency, we incorporate conditional adversarial learning into the editing process. This strategy helps to refine the edited results and maintain consistency throughout the animation. By integrating these methods, our GaussianAvatar-Editor achieves photorealistic and consistent results in animatable 4D Gaussian editing. We conduct comprehensive experiments across various subjects to validate the effectiveness of our proposed techniques, which demonstrates the superiority of our approach over existing methods. More results and code are available at: [Project Link](https://xiangyueliu.github.io/GaussianAvatar-Editor/).
comment: Accepted to 3DV 2025. [Project Link](https://xiangyueliu.github.io/GaussianAvatar-Editor/)
☆ Explainable artificial intelligence (XAI): from inherent explainability to large language models
Artificial Intelligence (AI) has continued to achieve tremendous success in recent times. However, the decision logic of these frameworks is often not transparent, making it difficult for stakeholders to understand, interpret or explain their behavior. This limitation hinders trust in machine learning systems and causes a general reluctance towards their adoption in practical applications, particularly in mission-critical domains like healthcare and autonomous driving. Explainable AI (XAI) techniques facilitate the explainability or interpretability of machine learning models, enabling users to discern the basis of the decision and possibly avert undesirable behavior. This comprehensive survey details the advancements of explainable AI methods, from inherently interpretable models to modern approaches for achieving interpretability of various black box models, including large language models (LLMs). Additionally, we review explainable AI techniques that leverage LLM and vision-language model (VLM) frameworks to automate or improve the explainability of other machine learning models. The use of LLM and VLM as interpretability methods particularly enables high-level, semantically meaningful explanations of model decisions and behavior. Throughout the paper, we highlight the scientific principles, strengths and weaknesses of state-of-the-art methods and outline different areas of improvement. Where appropriate, we also present qualitative and quantitative comparison results of various methods to show how they compare. Finally, we discuss the key challenges of XAI and directions for future research.
☆ Discrete Prior-based Temporal-coherent Content Prediction for Blind Face Video Restoration
Blind face video restoration aims to restore high-fidelity details from videos subjected to complex and unknown degradations. This task poses a significant challenge of managing temporal heterogeneity while at the same time maintaining stable face attributes. In this paper, we introduce a Discrete Prior-based Temporal-Coherent content prediction transformer to address the challenge, and our model is referred to as DP-TempCoh. Specifically, we incorporate a spatial-temporal-aware content prediction module to synthesize high-quality content from discrete visual priors, conditioned on degraded video tokens. To further enhance the temporal coherence of the predicted content, a motion statistics modulation module is designed to adjust the content, based on discrete motion priors in terms of cross-frame mean and variance. As a result, the statistics of the predicted content can match with that of real videos over time. By performing extensive experiments, we verify the effectiveness of the design elements and demonstrate the superior performance of our DP-TempCoh in both synthetically and naturally degraded video restoration.
☆ Surface-SOS: Self-Supervised Object Segmentation via Neural Surface Representation
Self-supervised Object Segmentation (SOS) aims to segment objects without any annotations. Under conditions of multi-camera inputs, the structural, textural and geometrical consistency among each view can be leveraged to achieve fine-grained object segmentation. To make better use of the above information, we propose Surface representation based Self-supervised Object Segmentation (Surface-SOS), a new framework to segment objects for each view by 3D surface representation from multi-view images of a scene. To model high-quality geometry surfaces for complex scenes, we design a novel scene representation scheme, which decomposes the scene into two complementary neural representation modules respectively with a Signed Distance Function (SDF). Moreover, Surface-SOS is able to refine single-view segmentation with multi-view unlabeled images, by introducing coarse segmentation masks as additional input. To the best of our knowledge, Surface-SOS is the first self-supervised approach that leverages neural surface representation to break the dependence on large amounts of annotated data and strong constraints. These constraints typically involve observing target objects against a static background or relying on temporal supervision in videos. Extensive experiments on standard benchmarks including LLFF, CO3D, BlendedMVS, TUM and several real-world scenes show that Surface-SOS always yields finer object masks than its NeRF-based counterparts and surpasses supervised single-view baselines remarkably. Code is available at: https://github.com/zhengxyun/Surface-SOS.
comment: Accepted by TIP
☆ A Multi-Scale Feature Extraction and Fusion Deep Learning Method for Classification of Wheat Diseases
Wheat is an important source of dietary fiber and protein that is negatively impacted by a number of risks to its growth. The difficulty of identifying and classifying wheat diseases is discussed with an emphasis on wheat loose smut, leaf rust, and crown and root rot. Addressing conditions like crown and root rot, this study introduces an innovative approach that integrates multi-scale feature extraction with advanced image segmentation techniques to enhance classification accuracy. The proposed method uses neural network models Xception, Inception V3, and ResNet 50 to train on a large wheat disease classification dataset 2020 in conjunction with an ensemble of machine vision classifiers, including voting and stacking. The study shows that the suggested methodology has a superior accuracy of 99.75% in the classification of wheat diseases when compared to current state-of-the-art approaches. A deep learning ensemble model Xception showed the highest accuracy.
☆ Physics-informed DeepCT: Sinogram Wavelet Decomposition Meets Masked Diffusion
Diffusion model shows remarkable potential on sparse-view computed tomography (SVCT) reconstruction. However, when a network is trained on a limited sample space, its generalization capability may be constrained, which degrades performance on unfamiliar data. For image generation tasks, this can lead to issues such as blurry details and inconsistencies between regions. To alleviate this problem, we propose a Sinogram-based Wavelet random decomposition And Random mask diffusion Model (SWARM) for SVCT reconstruction. Specifically, introducing a random mask strategy in the sinogram effectively expands the limited training sample space. This enables the model to learn a broader range of data distributions, enhancing its understanding and generalization of data uncertainty. In addition, applying a random training strategy to the high-frequency components of the sinogram wavelet enhances feature representation and improves the ability to capture details in different frequency bands, thereby improving performance and robustness. Two-stage iterative reconstruction method is adopted to ensure the global consistency of the reconstructed image while refining its details. Experimental results demonstrate that SWARM outperforms competing approaches in both quantitative and qualitative performance across various datasets.
☆ IE-Bench: Advancing the Measurement of Text-Driven Image Editing for Human Perception Alignment
Recent advances in text-driven image editing have been significant, yet the task of accurately evaluating these edited images continues to pose a considerable challenge. Different from the assessment of text-driven image generation, text-driven image editing is characterized by simultaneously conditioning on both text and a source image. The edited images often retain an intrinsic connection to the original image, which dynamically change with the semantics of the text. However, previous methods tend to solely focus on text-image alignment or have not aligned with human perception. In this work, we introduce the Text-driven Image Editing Benchmark suite (IE-Bench) to enhance the assessment of text-driven edited images. IE-Bench includes a database contains diverse source images, various editing prompts and the corresponding results different editing methods, and total 3,010 Mean Opinion Scores (MOS) provided by 25 human subjects. Furthermore, we introduce IE-QA, a multi-modality source-aware quality assessment method for text-driven image editing. To the best of our knowledge, IE-Bench offers the first IQA dataset and model tailored for text-driven image editing. Extensive experiments demonstrate IE-QA's superior subjective-alignments on the text-driven image editing task compared with previous metrics. We will make all related data and code available to the public.
☆ ForestProtector: An IoT Architecture Integrating Machine Vision and Deep Reinforcement Learning for Efficient Wildfire Monitoring
Early detection of forest fires is crucial to minimizing the environmental and socioeconomic damage they cause. Indeed, a fire's duration directly correlates with the difficulty and cost of extinguishing it. For instance, a fire burning for 1 minute might require 1 liter of water to extinguish, while a 2-minute fire could demand 100 liters, and a 10-minute fire might necessitate 1,000 liters. On the other hand, existing fire detection systems based on novel technologies (e.g., remote sensing, PTZ cameras, UAVs) are often expensive and require human intervention, making continuous monitoring of large areas impractical. To address this challenge, this work proposes a low-cost forest fire detection system that utilizes a central gateway device with computer vision capabilities to monitor a 360{\deg} field of view for smoke at long distances. A deep reinforcement learning agent enhances surveillance by dynamically controlling the camera's orientation, leveraging real-time sensor data (smoke levels, ambient temperature, and humidity) from distributed IoT devices. This approach enables automated wildfire monitoring across expansive areas while reducing false positives.
comment: Accepted for publication in the proceedings of the 11th International Conference on Automation, Robotics, and Applications (ICARA 2025)
☆ TalkingEyes: Pluralistic Speech-Driven 3D Eye Gaze Animation
Although significant progress has been made in the field of speech-driven 3D facial animation recently, the speech-driven animation of an indispensable facial component, eye gaze, has been overlooked by recent research. This is primarily due to the weak correlation between speech and eye gaze, as well as the scarcity of audio-gaze data, making it very challenging to generate 3D eye gaze motion from speech alone. In this paper, we propose a novel data-driven method which can generate diverse 3D eye gaze motions in harmony with the speech. To achieve this, we firstly construct an audio-gaze dataset that contains about 14 hours of audio-mesh sequences featuring high-quality eye gaze motion, head motion and facial motion simultaneously. The motion data is acquired by performing lightweight eye gaze fitting and face reconstruction on videos from existing audio-visual datasets. We then tailor a novel speech-to-motion translation framework in which the head motions and eye gaze motions are jointly generated from speech but are modeled in two separate latent spaces. This design stems from the physiological knowledge that the rotation range of eyeballs is less than that of head. Through mapping the speech embedding into the two latent spaces, the difficulty in modeling the weak correlation between speech and non-verbal motion is thus attenuated. Finally, our TalkingEyes, integrated with a speech-driven 3D facial motion generator, can synthesize eye gaze motion, eye blinks, head motion and facial motion collectively from speech. Extensive quantitative and qualitative evaluations demonstrate the superiority of the proposed method in generating diverse and natural 3D eye gaze motions from speech. The project page of this paper is: https://lkjkjoiuiu.github.io/TalkingEyes_Home/
☆ SLIM: Sim-to-Real Legged Instructive Manipulation via Long-Horizon Visuomotor Learning
We present a low-cost quadruped manipulation system that solves long-horizon real-world tasks, trained by reinforcement learning purely in simulation. The system comprises 1) a hierarchical design of a high-level policy for visual-mobile manipulation following instructions, and a low-level policy for quadruped movement and limb-control, 2) a progressive policy expansion approach for solving the long-horizon task together with a teacher-student framework for efficient high-level training of the high-level visuomotor policy, and 3) a suite of techniques for minimizing sim-to-real gaps. With budget-friendly but limited reliability and performance hardware, and just one wrist-mounted RGB camera, the entire system fully trained in simulation achieves high success rates for long horizon tasks involving search, move, grasp, and drop-into, with fluid sim-to-real transfer in a wide variety of indoor and outdoor scenes and lighting conditions.Extensive real-world evaluations show that on the long horizon mobile manipulation tasks, our system achieves good performance when transferred to real both in terms of task success rate and execution efficiency. Finally, we discuss the necessity of our sim-to-real techniques for legged mobile manipulation, and show their ablation performance.
☆ FoundationStereo: Zero-Shot Stereo Matching
Tremendous progress has been made in deep stereo matching to excel on benchmark datasets through per-domain fine-tuning. However, achieving strong zero-shot generalization - a hallmark of foundation models in other computer vision tasks - remains challenging for stereo matching. We introduce FoundationStereo, a foundation model for stereo depth estimation designed to achieve strong zero-shot generalization. To this end, we first construct a large-scale (1M stereo pairs) synthetic training dataset featuring large diversity and high photorealism, followed by an automatic self-curation pipeline to remove ambiguous samples. We then design a number of network architecture components to enhance scalability, including a side-tuning feature backbone that adapts rich monocular priors from vision foundation models to mitigate the sim-to-real gap, and long-range context reasoning for effective cost volume filtering. Together, these components lead to strong robustness and accuracy across domains, establishing a new standard in zero-shot stereo depth estimation.
☆ FLORA: Formal Language Model Enables Robust Training-free Zero-shot Object Referring Analysis
Object Referring Analysis (ORA), commonly known as referring expression comprehension, requires the identification and localization of specific objects in an image based on natural descriptions. Unlike generic object detection, ORA requires both accurate language understanding and precise visual localization, making it inherently more complex. Although recent pre-trained large visual grounding detectors have achieved significant progress, they heavily rely on extensively labeled data and time-consuming learning. To address these, we introduce a novel, training-free framework for zero-shot ORA, termed FLORA (Formal Language for Object Referring and Analysis). FLORA harnesses the inherent reasoning capabilities of large language models (LLMs) and integrates a formal language model - a logical framework that regulates language within structured, rule-based descriptions - to provide effective zero-shot ORA. More specifically, our formal language model (FLM) enables an effective, logic-driven interpretation of object descriptions without necessitating any training processes. Built upon FLM-regulated LLM outputs, we further devise a Bayesian inference framework and employ appropriate off-the-shelf interpretive models to finalize the reasoning, delivering favorable robustness against LLM hallucinations and compelling ORA performance in a training-free manner. In practice, our FLORA boosts the zero-shot performance of existing pretrained grounding detectors by up to around 45%. Our comprehensive evaluation across different challenging datasets also confirms that FLORA consistently surpasses current state-of-the-art zero-shot methods in both detection and segmentation tasks associated with zero-shot ORA. We believe our probabilistic parsing and reasoning of the LLM outputs elevate the reliability and interpretability of zero-shot ORA. We shall release codes upon publication.
♻ ☆ MVTamperBench: Evaluating Robustness of Vision-Language Models
Multimodal Large Language Models (MLLMs) have driven major advances in video understanding, yet their vulnerability to adversarial tampering and manipulations remains underexplored. To address this gap, we introduce MVTamperBench, a benchmark that systematically evaluates MLLM robustness against five prevalent tampering techniques: rotation, masking, substitution, repetition, and dropping. Built from 3.4K original videos-expanded to over 17K tampered clips spanning 19 video tasks. MVTamperBench challenges models to detect manipulations in spatial and temporal coherence. We evaluate 45 recent MLLMs from 15+ model families, revealing substantial variability in resilience across tampering types and showing that larger parameter counts do not necessarily guarantee robustness. MVTamperBench sets a new benchmark for developing tamper-resilient MLLM in safety-critical applications, including detecting clickbait, preventing harmful content distribution, and enforcing policies on media platforms. We release all code and data to foster open research in trustworthy video understanding. Code: https://amitbcp.github.io/MVTamperBench/ Data: https://huggingface.co/datasets/Srikant86/MVTamperBench
♻ ☆ Mesh2SLAM in VR: A Fast Geometry-Based SLAM Framework for Rapid Prototyping in Virtual Reality Applications
SLAM is a foundational technique with broad applications in robotics and AR/VR. SLAM simulations evaluate new concepts, but testing on resource-constrained devices, such as VR HMDs, faces challenges: high computational cost and restricted sensor data access. This work proposes a sparse framework using mesh geometry projections as features, which improves efficiency and circumvents direct sensor data access, advancing SLAM research as we demonstrate in VR and through numerical evaluation.
♻ ☆ ESVO2: Direct Visual-Inertial Odometry with Stereo Event Cameras
Event-based visual odometry is a specific branch of visual Simultaneous Localization and Mapping (SLAM) techniques, which aims at solving tracking and mapping subproblems (typically in parallel), by exploiting the special working principles of neuromorphic (i.e., event-based) cameras. Due to the motion-dependent nature of event data, explicit data association (i.e., feature matching) under large-baseline view-point changes is difficult to establish, making direct methods a more rational choice. However, state-of-the-art direct methods are limited by the high computational complexity of the mapping sub-problem and the degeneracy of camera pose tracking in certain degrees of freedom (DoF) in rotation. In this paper, we tackle these issues by building an event-based stereo visual-inertial odometry system on top of a direct pipeline. Specifically, to speed up the mapping operation, we propose an efficient strategy for sampling contour points according to the local dynamics of events. The mapping performance is also improved in terms of structure completeness and local smoothness by merging the temporal stereo and static stereo results. To circumvent the degeneracy of camera pose tracking in recovering the pitch and yaw components of general 6-DoF motion, we introduce IMU measurements as motion priors via pre-integration. To this end, a compact back-end is proposed for continuously updating the IMU bias and predicting the linear velocity, enabling an accurate motion prediction for camera pose tracking. The resulting system scales well with modern high-resolution event cameras and leads to better global positioning accuracy in large-scale outdoor environments. Extensive evaluations on five publicly available datasets featuring different resolutions and scenarios justify the superior performance of the proposed system against five state-of-the-art methods.
♻ ☆ BILTS: A Bi-Invariant Similarity Measure for Robust Object Trajectory Recognition under Reference Frame Variations
When similar object motions are performed in diverse contexts but are meant to be recognized under a single classification, these contextual variations act as disturbances that negatively affect accurate motion recognition. In this paper, we focus on contextual variations caused by reference frame variations. To robustly deal with these variations, similarity measures have been introduced that compare object motion trajectories in a context-invariant manner. However, most are highly sensitive to noise near singularities, where the measure is not uniquely defined, and lack bi-invariance (invariance to both world and body frame variations). To address these issues, we propose the novel \textit{Bi-Invariant Local Trajectory-Shape Similarity} (BILTS) measure. Compared to other measures, the BILTS measure uniquely offers bi-invariance, boundedness, and third-order shape identity. Aimed at practical implementations, we devised a discretized and regularized version of the BILTS measure which shows exceptional robustness to singularities. This is demonstrated through rigorous recognition experiments using multiple datasets. On average, BILTS attained the highest recognition ratio and least sensitivity to contextual variations compared to other invariant object motion similarity measures. We believe that the BILTS measure is a valuable tool for recognizing motions performed in diverse contexts and has potential in other applications, including the recognition, segmentation, and adaptation of both motion and force trajectories.
comment: This work has been submitted as a regular research paper for consideration in the Journal of Intelligent & Robotic Systems. The content in this preprint is identical to the version submitted for peer review, except for formatting differences required by the journal
♻ ☆ Bridging Diversity and Uncertainty in Active learning with Self-Supervised Pre-Training ICLR 2024
This study addresses the integration of diversity-based and uncertainty-based sampling strategies in active learning, particularly within the context of self-supervised pre-trained models. We introduce a straightforward heuristic called TCM that mitigates the cold start problem while maintaining strong performance across various data levels. By initially applying TypiClust for diversity sampling and subsequently transitioning to uncertainty sampling with Margin, our approach effectively combines the strengths of both strategies. Our experiments demonstrate that TCM consistently outperforms existing methods across various datasets in both low and high data regimes.
comment: Accepted at ICLR 2024 Workshop on Practical Machine Learning for Low Resource Settings (PML4LRS)
♻ ☆ Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models
We present Deep Compression Autoencoder (DC-AE), a new family of autoencoder models for accelerating high-resolution diffusion models. Existing autoencoder models have demonstrated impressive results at a moderate spatial compression ratio (e.g., 8x), but fail to maintain satisfactory reconstruction accuracy for high spatial compression ratios (e.g., 64x). We address this challenge by introducing two key techniques: (1) Residual Autoencoding, where we design our models to learn residuals based on the space-to-channel transformed features to alleviate the optimization difficulty of high spatial-compression autoencoders; (2) Decoupled High-Resolution Adaptation, an efficient decoupled three-phases training strategy for mitigating the generalization penalty of high spatial-compression autoencoders. With these designs, we improve the autoencoder's spatial compression ratio up to 128 while maintaining the reconstruction quality. Applying our DC-AE to latent diffusion models, we achieve significant speedup without accuracy drop. For example, on ImageNet 512x512, our DC-AE provides 19.1x inference speedup and 17.9x training speedup on H100 GPU for UViT-H while achieving a better FID, compared with the widely used SD-VAE-f8 autoencoder. Our code is available at https://github.com/mit-han-lab/efficientvit.
comment: Preprint. First two authors contributed equally to this work. Update: fix typo
♻ ☆ Generate E-commerce Product Background by Integrating Category Commonality and Personalized Style ICASSP 2025
The state-of-the-art methods for e-commerce product background generation suffer from the inefficiency of designing product-wise prompts when scaling up the production, as well as the ineffectiveness of describing fine-grained styles when customizing personalized backgrounds for some specific brands. To address these obstacles, we integrate the category commonality and personalized style into diffusion models. Concretely, we propose a Category-Wise Generator to enable large-scale background generation with only one model for the first time. A unique identifier in the prompt is assigned to each category, whose attention is located on the background by a mask-guided cross attention layer to learn the category-wise style. Furthermore, for products with specific and fine-grained requirements in layout, elements, etc, a Personality-Wise Generator is devised to learn such personalized style directly from a reference image to resolve textual ambiguities, and is trained in a self-supervised manner for more efficient training data usage. To advance research in this field, the first large-scale e-commerce product background generation dataset BG60k is constructed, which covers more than 60k product images from over 2k categories. Experiments demonstrate that our method could generate high-quality backgrounds for different categories, and maintain the personalized background style of reference images. BG60k will be available at \url{https://github.com/Whileherham/BG60k}.
comment: Accepted by ICASSP 2025
♻ ☆ LayerAnimate: Layer-specific Control for Animation
Animated video separates foreground and background elements into layers, with distinct processes for sketching, refining, coloring, and in-betweening. Existing video generation methods typically treat animation as a monolithic data domain, lacking fine-grained control over individual layers. In this paper, we introduce LayerAnimate, a novel architectural approach that enhances fine-grained control over individual animation layers within a video diffusion model, allowing users to independently manipulate foreground and background elements in distinct layers. To address the challenge of limited layer-specific data, we propose a data curation pipeline that features automated element segmentation, motion-state hierarchical merging, and motion coherence refinement. Through quantitative and qualitative comparisons, and user study, we demonstrate that LayerAnimate outperforms current methods in terms of animation quality, control precision, and usability, making it an ideal tool for both professional animators and amateur enthusiasts. This framework opens up new possibilities for layer-specific animation applications and creative flexibility. Our code is available at https://layeranimate.github.io.
comment: Project page: https://layeranimate.github.io
♻ ☆ A Survey on Deep Learning for Polyp Segmentation: Techniques, Challenges and Future Trends
Early detection and assessment of polyps play a crucial role in the prevention and treatment of colorectal cancer (CRC). Polyp segmentation provides an effective solution to assist clinicians in accurately locating and segmenting polyp regions. In the past, people often relied on manually extracted lower-level features such as color, texture, and shape, which often had issues capturing global context and lacked robustness to complex scenarios. With the advent of deep learning, more and more outstanding medical image segmentation algorithms based on deep learning networks have emerged, making significant progress in this field. This paper provides a comprehensive review of polyp segmentation algorithms. We first review some traditional algorithms based on manually extracted features and deep segmentation algorithms, then detail benchmark datasets related to the topic. Specifically, we carry out a comprehensive evaluation of recent deep learning models and results based on polyp sizes, considering the pain points of research topics and differences in network structures. Finally, we discuss the challenges of polyp segmentation and future trends in this field. The models, benchmark datasets, and source code links we collected are all published at https://github.com/taozh2017/Awesome-Polyp-Segmentation.
comment: Have been published in Visual Intelligence
♻ ☆ Isolated Diffusion: Optimizing Multi-Concept Text-to-Image Generation Training-Freely with Isolated Diffusion Guidance
Large-scale text-to-image diffusion models have achieved great success in synthesizing high-quality and diverse images given target text prompts. Despite the revolutionary image generation ability, current state-of-the-art models still struggle to deal with multi-concept generation accurately in many cases. This phenomenon is known as ``concept bleeding" and displays as the unexpected overlapping or merging of various concepts. This paper presents a general approach for text-to-image diffusion models to address the mutual interference between different subjects and their attachments in complex scenes, pursuing better text-image consistency. The core idea is to isolate the synthesizing processes of different concepts. We propose to bind each attachment to corresponding subjects separately with split text prompts. Besides, we introduce a revision method to fix the concept bleeding problem in multi-subject synthesis. We first depend on pre-trained object detection and segmentation models to obtain the layouts of subjects. Then we isolate and resynthesize each subject individually with corresponding text prompts to avoid mutual interference. Overall, we achieve a training-free strategy, named Isolated Diffusion, to optimize multi-concept text-to-image synthesis. It is compatible with the latest Stable Diffusion XL (SDXL) and prior Stable Diffusion (SD) models. We compare our approach with alternative methods using a variety of multi-concept text prompts and demonstrate its effectiveness with clear advantages in text-image consistency and user study.
comment: Accepted by IEEE Transactions on Visualization and Computer Graphics
♻ ☆ Expression Prompt Collaboration Transformer for Universal Referring Video Object Segmentation
Audio-guided Video Object Segmentation (A-VOS) and Referring Video Object Segmentation (R-VOS) are two highly related tasks that both aim to segment specific objects from video sequences according to expression prompts. However, due to the challenges of modeling representations for different modalities, existing methods struggle to strike a balance between interaction flexibility and localization precision. In this paper, we address this problem from two perspectives: the alignment of audio and text and the deep interaction among audio, text, and visual modalities. First, we propose a universal architecture, the Expression Prompt Collaboration Transformer, herein EPCFormer. Next, we propose an Expression Alignment (EA) mechanism for audio and text. The proposed EPCFormer exploits the fact that audio and text prompts referring to the same objects are semantically equivalent by using contrastive learning for both types of expressions. Then, to facilitate deep interactions among audio, text, and visual modalities, we introduce an Expression-Visual Attention (EVA) module. The knowledge of video object segmentation in terms of the expression prompts can seamlessly transfer between the two tasks by deeply exploring complementary cues between text and audio. Experiments on well-recognized benchmarks demonstrate that our EPCFormer attains state-of-the-art results on both tasks. The source code will be made publicly available at https://github.com/lab206/EPCFormer.
comment: Accepted to Knowledge-Based Systems (KBS). The source code will be made publicly available at https://github.com/lab206/EPCFormer
♻ ☆ Tarsier2: Advancing Large Vision-Language Models from Detailed Video Description to Comprehensive Video Understanding
We introduce Tarsier2, a state-of-the-art large vision-language model (LVLM) designed for generating detailed and accurate video descriptions, while also exhibiting superior general video understanding capabilities. Tarsier2 achieves significant advancements through three key upgrades: (1) Scaling pre-training data from 11M to 40M video-text pairs, enriching both volume and diversity; (2) Performing fine-grained temporal alignment during supervised fine-tuning; (3) Using model-based sampling to automatically construct preference data and applying DPO training for optimization. Extensive experiments show that Tarsier2-7B consistently outperforms leading proprietary models, including GPT-4o and Gemini 1.5 Pro, in detailed video description tasks. On the DREAM-1K benchmark, Tarsier2-7B improves F1 by 2.8\% over GPT-4o and 5.8\% over Gemini-1.5-Pro. In human side-by-side evaluations, Tarsier2-7B shows a +8.6\% performance advantage over GPT-4o and +24.9\% over Gemini-1.5-Pro. Tarsier2-7B also sets new state-of-the-art results across 15 public benchmarks, spanning tasks such as video question-answering, video grounding, hallucination test, and embodied question-answering, demonstrating its versatility as a robust generalist vision-language model.
♻ ☆ Continuous Urban Change Detection from Satellite Image Time Series with Temporal Feature Refinement and Multi-Task Integration
Urbanization advances at unprecedented rates, resulting in negative effects on the environment and human well-being. Remote sensing has the potential to mitigate these effects by supporting sustainable development strategies with accurate information on urban growth. Deep learning-based methods have achieved promising urban change detection results from optical satellite image pairs using convolutional neural networks (ConvNets), transformers, and a multi-task learning setup. However, transformers have not been leveraged for urban change detection with multi-temporal data, i.e., >2 images, and multi-task learning methods lack integration approaches that combine change and segmentation outputs. To fill this research gap, we propose a continuous urban change detection method that identifies changes in each consecutive image pair of a satellite image time series (SITS). Specifically, we propose a temporal feature refinement (TFR) module that utilizes self-attention to improve ConvNet-based multi-temporal building representations. Furthermore, we propose a multi-task integration (MTI) module that utilizes Markov networks to find an optimal building map time series based on segmentation and dense change outputs. The proposed method effectively identifies urban changes based on high-resolution SITS acquired by the PlanetScope constellation (F1 score 0.551) and Gaofen-2 (F1 score 0.440). Moreover, our experiments on two challenging datasets demonstrate the effectiveness of the proposed method compared to bi-temporal and multi-temporal urban change detection and segmentation methods.
comment: Under review at IEEE Transactions on Geoscience and Remote Sensing, Code will be available at https://github.com/SebastianHafner/ContUrbanCD.git
♻ ☆ Mamba2D: A Natively Multi-Dimensional State-Space Model for Vision Tasks
State-Space Models (SSMs) have recently emerged as a powerful and efficient alternative to the long-standing transformer architecture. However, existing SSM conceptualizations retain deeply rooted biases from their roots in natural language processing. This constrains their ability to appropriately model the spatially-dependent characteristics of visual inputs. In this paper, we address these limitations by re-deriving modern selective state-space techniques, starting from a natively multidimensional formulation. Currently, prior works attempt to apply natively 1D SSMs to 2D data (i.e. images) by relying on arbitrary combinations of 1D scan directions to capture spatial dependencies. In contrast, Mamba2D improves upon this with a single 2D scan direction that factors in both dimensions of the input natively, effectively modelling spatial dependencies when constructing hidden states. Mamba2D shows comparable performance to prior adaptations of SSMs for vision tasks, on standard image classification evaluations with the ImageNet-1K dataset. Source code is available at https://github.com/cocoalex00/Mamba2D.
♻ ☆ Model Synthesis for Zero-Shot Model Attribution
Nowadays, generative models are shaping various fields such as art, design, and human-computer interaction, yet accompanied by challenges related to copyright infringement and content management. In response, existing research seeks to identify the unique fingerprints on the images they generate, which can be leveraged to attribute the generated images to their source models. Existing methods, however, are constrained to identifying models within a static set included in the classifier training, failing to adapt to newly emerged unseen models dynamically. To bridge this gap, we aim to develop a generalized model fingerprint extractor capable of zero-shot attribution, effectively attributes unseen models without exposure during training. Central to our method is a model synthesis technique, which generates numerous synthetic models mimicking the fingerprint patterns of real-world generative models. The design of the synthesis technique is motivated by observations on how the basic generative model's architecture building blocks and parameters influence fingerprint patterns, and it is validated through two designed metrics that examine synthetic models' fidelity and diversity. Our experiments demonstrate that this fingerprint extractor, trained solely on synthetic models, achieves impressive zero-shot generalization on a wide range of real-world generative models, improving model identification and verification accuracy on unseen models by over 40% and 15%, respectively, compared to existing approaches.
comment: under review
♻ ☆ Multi-stage Deep Learning Artifact Reduction for Pallel-beam Computed Tomography
Computed Tomography (CT) using synchrotron radiation is a powerful technique that, compared to lab-CT techniques, boosts high spatial and temporal resolution while also providing access to a range of contrast-formation mechanisms. The acquired projection data is typically processed by a computational pipeline composed of multiple stages. Artifacts introduced during data acquisition can propagate through the pipeline, and degrade image quality in the reconstructed images. Recently, deep learning has shown significant promise in enhancing image quality for images representing scientific data. This success has driven increasing adoption of deep learning techniques in CT imaging. Various approaches have been proposed to incorporate deep learning into computational pipelines, but each has limitations in addressing artifacts effectively and efficiently in synchrotron CT, either in properly addressing the specific artifacts, or in computational efficiency. Recognizing these challenges, we introduce a novel method that incorporates separate deep learning models at each stage of the tomography pipeline-projection, sinogram, and reconstruction-to address specific artifacts locally in a data-driven way. Our approach includes bypass connections that feed both the outputs from previous stages and raw data to subsequent stages, minimizing the risk of error propagation. Extensive evaluations on both simulated and real-world datasets illustrate that our approach effectively reduces artifacts and outperforms comparison methods.
♻ ☆ IncSAR: A Dual Fusion Incremental Learning Framework for SAR Target Recognition
Deep learning techniques have achieved significant success in Synthetic Aperture Radar (SAR) target recognition using predefined datasets in static scenarios. However, real-world applications demand that models incrementally learn new information without forgetting previously acquired knowledge. The challenge of catastrophic forgetting, where models lose past knowledge when adapting to new tasks, remains a critical issue. In this paper, we introduce IncSAR, an incremental learning framework designed to tackle catastrophic forgetting in SAR target recognition. IncSAR combines the power of a Vision Transformer (ViT) and a custom-designed Convolutional Neural Network (CNN) in a dual-branch architecture, integrated via a late-fusion strategy. Additionally, we explore the use of TinyViT to reduce computational complexity and propose an attention mechanism to dynamically enhance feature representation. To mitigate the speckle noise inherent in SAR images, we employ a denoising module based on a neural network approximation of Robust Principal Component Analysis (RPCA), leveraging a simple neural network for efficient noise reduction in SAR imagery. Moreover, a random projection layer improves the linear separability of features, and a variant of Linear Discriminant Analysis (LDA) decorrelates extracted class prototypes for better generalization. Extensive experiments on the MSTAR, SAR-AIRcraft-1.0, and OpenSARShip benchmark datasets demonstrate that IncSAR significantly outperforms state-of-the-art approaches, achieving a 99.63\% average accuracy and a 0.33\% performance drop, representing an 89\% improvement in retention compared to existing techniques. The source code is available at https://github.com/geokarant/IncSAR.
♻ ☆ VLSBench: Unveiling Visual Leakage in Multimodal Safety
Safety concerns of Multimodal large language models (MLLMs) have gradually become an important problem in various applications. Surprisingly, previous works indicate a counter-intuitive phenomenon that using textual unlearning to align MLLMs achieves comparable safety performances with MLLMs trained with image-text pairs. To explain such a counter-intuitive phenomenon, we discover a visual safety information leakage (VSIL) problem in existing multimodal safety benchmarks, i.e., the potentially risky and sensitive content in the image has been revealed in the textual query. In this way, MLLMs can easily refuse these sensitive text-image queries according to textual queries. However, image-text pairs without VSIL are common in real-world scenarios and are overlooked by existing multimodal safety benchmarks. To this end, we construct multimodal visual leakless safety benchmark (VLSBench) preventing visual safety leakage from image to textual query with 2.4k image-text pairs. Experimental results indicate that VLSBench poses a significant challenge to both open-source and close-source MLLMs, including LLaVA, Qwen2-VL, Llama3.2-Vision, and GPT-4o. This study demonstrates that textual alignment is enough for multimodal safety scenarios with VSIL, while multimodal alignment is a more promising solution for multimodal safety scenarios without VSIL. Please see our code and data at: https://hxhcreate.github.io/vlsbench.github.io/
♻ ☆ SARATR-X: Towards Building A Foundation Model for SAR Target Recognition
Despite the remarkable progress in synthetic aperture radar automatic target recognition (SAR ATR), recent efforts have concentrated on detecting and classifying a specific category, e.g., vehicles, ships, airplanes, or buildings. One of the fundamental limitations of the top-performing SAR ATR methods is that the learning paradigm is supervised, task-specific, limited-category, closed-world learning, which depends on massive amounts of accurately annotated samples that are expensively labeled by expert SAR analysts and have limited generalization capability and scalability. In this work, we make the first attempt towards building a foundation model for SAR ATR, termed SARATR-X. SARATR-X learns generalizable representations via self-supervised learning (SSL) and provides a cornerstone for label-efficient model adaptation to generic SAR target detection and classification tasks. Specifically, SARATR-X is trained on 0.18 M unlabelled SAR target samples, which are curated by combining contemporary benchmarks and constitute the largest publicly available dataset till now. Considering the characteristics of SAR images, a backbone tailored for SAR ATR is carefully designed, and a two-step SSL method endowed with multi-scale gradient features was applied to ensure the feature diversity and model scalability of SARATR-X. The capabilities of SARATR-X are evaluated on classification under few-shot and robustness settings and detection across various categories and scenes, and impressive performance is achieved, often competitive with or even superior to prior fully supervised, semi-supervised, or self-supervised algorithms. Our SARATR-X and the curated dataset are released at https://github.com/waterdisappear/SARATR-X to foster research into foundation models for SAR image interpretation.
comment: 20 pages, 9 figures
♻ ☆ Mitigating analytical variability in fMRI results with style transfer
We propose a novel approach to improve the reproducibility of neuroimaging results by converting statistic maps across different functional MRI pipelines. We make the assumption that pipelines used to compute fMRI statistic maps can be considered as a style component and we propose to use different generative models, among which, Generative Adversarial Networks (GAN) and Diffusion Models (DM) to convert statistic maps across different pipelines. We explore the performance of multiple GAN frameworks, and design a new DM framework for unsupervised multi-domain styletransfer. We constrain the generation of 3D fMRI statistic maps using the latent space of an auxiliary classifier that distinguishes statistic maps from different pipelines and extend traditional sampling techniques used in DM to improve the transition performance. Our experiments demonstrate that our proposed methods aresuccessful: pipelines can indeed be transferred as a style component, providing animportant source of data augmentation for future medical studies.
♻ ☆ Accelerating lensed quasars discovery and modeling with physics-informed variational autoencoders
Strongly lensed quasars provide valuable insights into the rate of cosmic expansion, the distribution of dark matter in foreground deflectors, and the characteristics of quasar hosts. However, detecting them in astronomical images is difficult due to the prevalence of non-lensing objects. To address this challenge, we developed a generative deep learning model called VariLens, built upon a physics-informed variational autoencoder. This model seamlessly integrates three essential modules: image reconstruction, object classification, and lens modeling, offering a fast and comprehensive approach to strong lens analysis. VariLens is capable of rapidly determining both (1) the probability that an object is a lens system and (2) key parameters of a singular isothermal ellipsoid (SIE) mass model -- including the Einstein radius ($\theta_\mathrm{E}$), lens center, and ellipticity -- in just milliseconds using a single CPU. A direct comparison of VariLens estimates with traditional lens modeling for 20 known lensed quasars within the Subaru Hyper Suprime-Cam (HSC) footprint shows good agreement, with both results consistent within $2\sigma$ for systems with $\theta_\mathrm{E}<3$ arcsecs. To identify new lensed quasar candidates, we begin with an initial sample of approximately 80 million sources, combining HSC data with multiwavelength information from various surveys. After applying a photometric preselection aimed at locating $z>1.5$ sources, the number of candidates is reduced to 710,966. Subsequently, VariLens highlights 13,831 sources, each showing a high likelihood of being a lens. A visual assessment of these objects results in 42 promising candidates that await spectroscopic confirmation. These results underscore the potential of automated deep learning pipelines to efficiently detect and model strong lenses in large datasets.
comment: Submitted to the Astronomy & Astrophysics journal and updated to reflect the revised version. The paper consists of 17 main pages, 14 figures, and 5 tables. We welcome feedback and comments from readers!
♻ ☆ WaveDH: Wavelet Sub-bands Guided ConvNet for Efficient Image Dehazing
The surge in interest regarding image dehazing has led to notable advancements in deep learning-based single image dehazing approaches, exhibiting impressive performance in recent studies. Despite these strides, many existing methods fall short in meeting the efficiency demands of practical applications. In this paper, we introduce WaveDH, a novel and compact ConvNet designed to address this efficiency gap in image dehazing. Our WaveDH leverages wavelet sub-bands for guided up-and-downsampling and frequency-aware feature refinement. The key idea lies in utilizing wavelet decomposition to extract low-and-high frequency components from feature levels, allowing for faster processing while upholding high-quality reconstruction. The downsampling block employs a novel squeeze-and-attention scheme to optimize the feature downsampling process in a structurally compact manner through wavelet domain learning, preserving discriminative features while discarding noise components. In our upsampling block, we introduce a dual-upsample and fusion mechanism to enhance high-frequency component awareness, aiding in the reconstruction of high-frequency details. Departing from conventional dehazing methods that treat low-and-high frequency components equally, our feature refinement block strategically processes features with a frequency-aware approach. By employing a coarse-to-fine methodology, it not only refines the details at frequency levels but also significantly optimizes computational costs. The refinement is performed in a maximum 8x downsampled feature space, striking a favorable efficiency-vs-accuracy trade-off. Extensive experiments demonstrate that our method, WaveDH, outperforms many state-of-the-art methods on several image dehazing benchmarks with significantly reduced computational costs. Our code is available at https://github.com/AwesomeHwang/WaveDH.
comment: Under Review
♻ ☆ Text-guided Image Restoration and Semantic Enhancement for Text-to-Image Person Retrieval
The goal of Text-to-Image Person Retrieval (TIPR) is to retrieve specific person images according to the given textual descriptions. A primary challenge in this task is bridging the substantial representational gap between visual and textual modalities. The prevailing methods map texts and images into unified embedding space for matching, while the intricate semantic correspondences between texts and images are still not effectively constructed. To address this issue, we propose a novel TIPR framework to build fine-grained interactions and alignment between person images and the corresponding texts. Specifically, via fine-tuning the Contrastive Language-Image Pre-training (CLIP) model, a visual-textual dual encoder is firstly constructed, to preliminarily align the image and text features. Secondly, a Text-guided Image Restoration (TIR) auxiliary task is proposed to map abstract textual entities to specific image regions, improving the alignment between local textual and visual embeddings. Additionally, a cross-modal triplet loss is presented to handle hard samples, and further enhance the model's discriminability for minor differences. Moreover, a pruning-based text data augmentation approach is proposed to enhance focus on essential elements in descriptions, thereby avoiding excessive model attention to less significant information. The experimental results show our proposed method outperforms state-of-the-art methods on three popular benchmark datasets, and the code will be made publicly available at https://github.com/Delong-liu-bupt/SEN.
comment: The paper was withdrawn due to a dispute among the authors regarding the content of the article
♻ ☆ Text-guided Synthetic Geometric Augmentation for Zero-shot 3D Understanding
Zero-shot recognition models require extensive training data for generalization. However, in zero-shot 3D classification, collecting 3D data and captions is costly and laborintensive, posing a significant barrier compared to 2D vision. Recent advances in generative models have achieved unprecedented realism in synthetic data production, and recent research shows the potential for using generated data as training data. Here, naturally raising the question: Can synthetic 3D data generated by generative models be used as expanding limited 3D datasets? In response, we present a synthetic 3D dataset expansion method, Textguided Geometric Augmentation (TeGA). TeGA is tailored for language-image-3D pretraining, which achieves SoTA in zero-shot 3D classification, and uses a generative textto-3D model to enhance and extend limited 3D datasets. Specifically, we automatically generate text-guided synthetic 3D data and introduce a consistency filtering strategy to discard noisy samples where semantics and geometric shapes do not match with text. In the experiment to double the original dataset size using TeGA, our approach demonstrates improvements over the baselines, achieving zeroshot performance gains of 3.0% on Objaverse-LVIS, 4.6% on ScanObjectNN, and 8.7% on ModelNet40. These results demonstrate that TeGA effectively bridges the 3D data gap, enabling robust zero-shot 3D classification even with limited real training data and paving the way for zero-shot 3D vision application.
♻ ☆ SuperNeRF-GAN: A Universal 3D-Consistent Super-Resolution Framework for Efficient and Enhanced 3D-Aware Image Synthesis
Neural volume rendering techniques, such as NeRF, have revolutionized 3D-aware image synthesis by enabling the generation of images of a single scene or object from various camera poses. However, the high computational cost of NeRF presents challenges for synthesizing high-resolution (HR) images. Most existing methods address this issue by leveraging 2D super-resolution, which compromise 3D-consistency. Other methods propose radiance manifolds or two-stage generation to achieve 3D-consistent HR synthesis, yet they are limited to specific synthesis tasks, reducing their universality. To tackle these challenges, we propose SuperNeRF-GAN, a universal framework for 3D-consistent super-resolution. A key highlight of SuperNeRF-GAN is its seamless integration with NeRF-based 3D-aware image synthesis methods and it can simultaneously enhance the resolution of generated images while preserving 3D-consistency and reducing computational cost. Specifically, given a pre-trained generator capable of producing a NeRF representation such as tri-plane, we first perform volume rendering to obtain a low-resolution image with corresponding depth and normal map. Then, we employ a NeRF Super-Resolution module which learns a network to obtain a high-resolution NeRF. Next, we propose a novel Depth-Guided Rendering process which contains three simple yet effective steps, including the construction of a boundary-correct multi-depth map through depth aggregation, a normal-guided depth super-resolution and a depth-guided NeRF rendering. Experimental results demonstrate the superior efficiency, 3D-consistency, and quality of our approach. Additionally, ablation studies confirm the effectiveness of our proposed components.
♻ ☆ DX2CT: Diffusion Model for 3D CT Reconstruction from Bi or Mono-planar 2D X-ray(s)
Computational tomography (CT) provides high-resolution medical imaging, but it can expose patients to high radiation. X-ray scanners have low radiation exposure, but their resolutions are low. This paper proposes a new conditional diffusion model, DX2CT, that reconstructs three-dimensional (3D) CT volumes from bi or mono-planar X-ray image(s). Proposed DX2CT consists of two key components: 1) modulating feature maps extracted from two-dimensional (2D) X-ray(s) with 3D positions of CT volume using a new transformer and 2) effectively using the modulated 3D position-aware feature maps as conditions of DX2CT. In particular, the proposed transformer can provide conditions with rich information of a target CT slice to the conditional diffusion model, enabling high-quality CT reconstruction. Our experiments with the bi or mono-planar X-ray(s) benchmark datasets show that proposed DX2CT outperforms several state-of-the-art methods. Our codes and model will be available at: https://www.github.com/intyeger/DX2CT.
♻ ☆ MoRe: Class Patch Attention Needs Regularization for Weakly Supervised Semantic Segmentation AAAI 2025
Weakly Supervised Semantic Segmentation (WSSS) with image-level labels typically uses Class Activation Maps (CAM) to achieve dense predictions. Recently, Vision Transformer (ViT) has provided an alternative to generate localization maps from class-patch attention. However, due to insufficient constraints on modeling such attention, we observe that the Localization Attention Maps (LAM) often struggle with the artifact issue, i.e., patch regions with minimal semantic relevance are falsely activated by class tokens. In this work, we propose MoRe to address this issue and further explore the potential of LAM. Our findings suggest that imposing additional regularization on class-patch attention is necessary. To this end, we first view the attention as a novel directed graph and propose the Graph Category Representation module to implicitly regularize the interaction among class-patch entities. It ensures that class tokens dynamically condense the related patch information and suppress unrelated artifacts at a graph level. Second, motivated by the observation that CAM from classification weights maintains smooth localization of objects, we devise the Localization-informed Regularization module to explicitly regularize the class-patch attention. It directly mines the token relations from CAM and further supervises the consistency between class and patch tokens in a learnable manner. Extensive experiments are conducted on PASCAL VOC and MS COCO, validating that MoRe effectively addresses the artifact issue and achieves state-of-the-art performance, surpassing recent single-stage and even multi-stage methods. Code is available at https://github.com/zwyang6/MoRe.
comment: AAAI 2025
♻ ☆ Elucidating the Design Space of Dataset Condensation NeurIPS 2024
Dataset condensation, a concept within data-centric learning, efficiently transfers critical attributes from an original dataset to a synthetic version, maintaining both diversity and realism. This approach significantly improves model training efficiency and is adaptable across multiple application areas. Previous methods in dataset condensation have faced challenges: some incur high computational costs which limit scalability to larger datasets (e.g., MTT, DREAM, and TESLA), while others are restricted to less optimal design spaces, which could hinder potential improvements, especially in smaller datasets (e.g., SRe2L, G-VBSM, and RDED). To address these limitations, we propose a comprehensive design framework that includes specific, effective strategies like implementing soft category-aware matching and adjusting the learning rate schedule. These strategies are grounded in empirical evidence and theoretical backing. Our resulting approach, Elucidate Dataset Condensation (EDC), establishes a benchmark for both small and large-scale dataset condensation. In our testing, EDC achieves state-of-the-art accuracy, reaching 48.6% on ImageNet-1k with a ResNet-18 model at an IPC of 10, which corresponds to a compression ratio of 0.78%. This performance exceeds those of SRe2L, G-VBSM, and RDED by margins of 27.3%, 17.2%, and 6.6%, respectively.
comment: Accepted by NeurIPS 2024
♻ ☆ Harnessing small projectors and multiple views for efficient vision pretraining NeurIPS 2024
Recent progress in self-supervised (SSL) visual representation learning has led to the development of several different proposed frameworks that rely on augmentations of images but use different loss functions. However, there are few theoretically grounded principles to guide practice, so practical implementation of each SSL framework requires several heuristics to achieve competitive performance. In this work, we build on recent analytical results to design practical recommendations for competitive and efficient SSL that are grounded in theory. Specifically, recent theory tells us that existing SSL frameworks are minimizing the same idealized loss, which is to learn features that best match the data similarity kernel defined by the augmentations used. We show how this idealized loss can be reformulated to a functionally equivalent loss that is more efficient to compute. We study the implicit bias of using gradient descent to minimize our reformulated loss function and find that using a stronger orthogonalization constraint with a reduced projector dimensionality should yield good representations. Furthermore, the theory tells us that approximating the reformulated loss should be improved by increasing the number of augmentations, and as such using multiple augmentations should lead to improved convergence. We empirically verify our findings on CIFAR, STL and Imagenet datasets, wherein we demonstrate an improved linear readout performance when training a ResNet-backbone using our theoretically grounded recommendations. Remarkably, we also demonstrate that by leveraging these insights, we can reduce the pretraining dataset size by up to 2$\times$ while maintaining downstream accuracy simply by using more data augmentations. Taken together, our work provides theoretically grounded recommendations that can be used to improve SSL convergence and efficiency.
comment: Accepted to NeurIPS 2024
♻ ☆ OPCap:Object-aware Prompting Captioning
In the field of image captioning, the phenomenon where missing or nonexistent objects are used to explain an image is referred to as object bias (or hallucination). To mitigate this issue, we propose a target-aware prompting strategy. This method first extracts object labels and their spatial information from the image using an object detector. Then, an attribute predictor further refines the semantic features of the objects. These refined features are subsequently integrated and fed into the decoder, enhancing the model's understanding of the image context. Experimental results on the COCO and nocaps datasets demonstrate that OPCap effectively mitigates hallucination and significantly improves the quality of generated captions.
♻ ☆ Driving in the Occupancy World: Vision-Centric 4D Occupancy Forecasting and Planning via World Models for Autonomous Driving AAAI2025
World models envision potential future states based on various ego actions. They embed extensive knowledge about the driving environment, facilitating safe and scalable autonomous driving. Most existing methods primarily focus on either data generation or the pretraining paradigms of world models. Unlike the aforementioned prior works, we propose Drive-OccWorld, which adapts a vision-centric 4D forecasting world model to end-to-end planning for autonomous driving. Specifically, we first introduce a semantic and motion-conditional normalization in the memory module, which accumulates semantic and dynamic information from historical BEV embeddings. These BEV features are then conveyed to the world decoder for future occupancy and flow forecasting, considering both geometry and spatiotemporal modeling. Additionally, we propose injecting flexible action conditions, such as velocity, steering angle, trajectory, and commands, into the world model to enable controllable generation and facilitate a broader range of downstream applications. Furthermore, we explore integrating the generative capabilities of the 4D world model with end-to-end planning, enabling continuous forecasting of future states and the selection of optimal trajectories using an occupancy-based cost function. Comprehensive experiments conducted on the nuScenes, nuScenes-Occupancy, and Lyft-Level5 datasets illustrate that our method can generate plausible and controllable 4D occupancy, paving the way for advancements in driving world generation and end-to-end planning. Project page: https://drive-occworld.github.io/
comment: Accepted by AAAI2025
♻ ☆ Deep Plug-and-Play HIO Approach for Phase Retrieval
In the phase retrieval problem, the aim is the recovery of an unknown image from intensity-only measurements such as Fourier intensity. Although there are several solution approaches, solving this problem is challenging due to its nonlinear and ill-posed nature. Recently, learning-based approaches have emerged as powerful alternatives to the analytical methods for several inverse problems. In the context of phase retrieval, a novel plug-and-play approach that exploits learning-based prior and efficient update steps has been presented at the Computational Optical Sensing and Imaging topical meeting, with demonstrated state-of-the-art performance. The key idea was to incorporate learning-based prior to the Gerchberg-Saxton type algorithms through plug-and-play regularization. In this paper, we present the mathematical development of the method including the derivation of its analytical update steps based on half-quadratic splitting and comparatively evaluate its performance through extensive simulations on a large test dataset. The results show the effectiveness of the method in terms of both image quality, computational efficiency, and robustness to initialization and noise.
comment: 16 pages, 5 figures
♻ ☆ Instruction-Guided Fusion of Multi-Layer Visual Features in Large Vision-Language Models
Large Vision-Language Models (LVLMs) have achieved remarkable success in a wide range of multimodal tasks by integrating pre-trained vision encoders and large language models. However, current LVLMs primarily rely on visual features extracted from the final layers of the vision encoder, overlooking the complementary information available in shallower layers. While recent approaches have explored the use of multilayer visual features in LVLMs, they tend to be task-agnostic and fail to examine the dependencies of hierarchical visual features on specific tasks. To address these gaps, we systematically investigate the contributions of visual features from different encoder layers using 18 benchmarks spanning 6 task categories. Our findings reveal that multilayer features provide complementary strengths with varying task dependencies, and uniform fusion leads to suboptimal performance. Building on these insights, we propose the instruction-guided vision aggregator, a module that dynamically integrates multi-layer visual features based on textual instructions, without increasing the number of visual tokens. Extensive evaluations demonstrate the superior performance of our method. Additionally, an in-depth analysis of the aggregator's behavior highlights the dominance of mid-to-high-level features in semantic-rich tasks and the critical role of low-level features in fine-grained perception.
♻ ☆ Myriad: Large Multimodal Model by Applying Vision Experts for Industrial Anomaly Detection
Due to the training configuration, traditional industrial anomaly detection (IAD) methods have to train a specific model for each deployment scenario, which is insufficient to meet the requirements of modern design and manufacturing. On the contrary, large multimodal models~(LMMs) have shown eminent generalization ability on various vision tasks, and their perception and comprehension capabilities imply the potential of applying LMMs on IAD tasks. However, we observe that even though the LMMs have abundant knowledge about industrial anomaly detection in the textual domain, the LMMs are unable to leverage the knowledge due to the modality gap between textual and visual domains. To stimulate the relevant knowledge in LMMs and adapt the LMMs towards anomaly detection tasks, we introduce existing IAD methods as vision experts and present a novel large multimodal model applying vision experts for industrial anomaly detection~(abbreviated to {Myriad}). Specifically, we utilize the anomaly map generated by the vision experts as guidance for LMMs, such that the vision model is guided to pay more attention to anomalous regions. Then, the visual features are modulated via an adapter to fit the anomaly detection tasks, which are fed into the language model together with the vision expert guidance and human instructions to generate the final outputs. Extensive experiments are applied on MVTec-AD, VisA, and PCB Bank benchmarks demonstrate that our proposed method not only performs favorably against state-of-the-art methods, but also inherits the flexibility and instruction-following ability of LMMs in the field of IAD. Source code and pre-trained models are publicly available at \url{https://github.com/tzjtatata/Myriad}.
comment: 8 pages, 7 figures
♻ ☆ TraceFL: Interpretability-Driven Debugging in Federated Learning via Neuron Provenance ICSE
In Federated Learning, clients train models on local data and send updates to a central server, which aggregates them into a global model using a fusion algorithm. This collaborative yet privacy-preserving training comes at a cost. FL developers face significant challenges in attributing global model predictions to specific clients. Localizing responsible clients is a crucial step towards (a) excluding clients primarily responsible for incorrect predictions and (b) encouraging clients who contributed high-quality models to continue participating in the future. Existing ML debugging approaches are inherently inapplicable as they are designed for single-model, centralized training. We introduce TraceFL, a fine-grained neuron provenance capturing mechanism that identifies clients responsible for a global model's prediction by tracking the flow of information from individual clients to the global model. Since inference on different inputs activates a different set of neurons of the global model, TraceFL dynamically quantifies the significance of the global model's neurons in a given prediction, identifying the most crucial neurons in the global model. It then maps them to the corresponding neurons in every participating client to determine each client's contribution, ultimately localizing the responsible client. We evaluate TraceFL on six datasets, including two real-world medical imaging datasets and four neural networks, including advanced models such as GPT. TraceFL achieves 99% accuracy in localizing the responsible client in FL tasks spanning both image and text classification tasks. At a time when state-of-the-artML debugging approaches are mostly domain-specific (e.g., image classification only), TraceFL is the first technique to enable highly accurate automated reasoning across a wide range of FL applications.
comment: Accepted at 2025 IEEE/ACM 47th International Conference on Software Engineering (ICSE)
♻ ☆ Epicardium Prompt-guided Real-time Cardiac Ultrasound Frame-to-volume Registration MICCAI 2024
A comprehensive guidance view for cardiac interventional surgery can be provided by the real-time fusion of the intraoperative 2D images and preoperative 3D volume based on the ultrasound frame-to-volume registration. However, cardiac ultrasound images are characterized by a low signal-to-noise ratio and small differences between adjacent frames, coupled with significant dimension variations between 2D frames and 3D volumes to be registered, resulting in real-time and accurate cardiac ultrasound frame-to-volume registration being a very challenging task. This paper introduces a lightweight end-to-end Cardiac Ultrasound frame-to-volume Registration network, termed CU-Reg. Specifically, the proposed model leverages epicardium prompt-guided anatomical clues to reinforce the interaction of 2D sparse and 3D dense features, followed by a voxel-wise local-global aggregation of enhanced features, thereby boosting the cross-dimensional matching effectiveness of low-quality ultrasound modalities. We further embed an inter-frame discriminative regularization term within the hybrid supervised learning to increase the distinction between adjacent slices in the same ultrasound volume to ensure registration stability. Experimental results on the reprocessed CAMUS dataset demonstrate that our CU-Reg surpasses existing methods in terms of registration accuracy and efficiency, meeting the guidance requirements of clinical cardiac interventional surgery.
comment: This paper has been accepted by MICCAI 2024
♻ ☆ NeuManifold: Neural Watertight Manifold Reconstruction with Efficient and High-Quality Rendering Support
We present a method for generating high-quality watertight manifold meshes from multi-view input images. Existing volumetric rendering methods are robust in optimization but tend to generate noisy meshes with poor topology. Differentiable rasterization-based methods can generate high-quality meshes but are sensitive to initialization. Our method combines the benefits of both worlds; we take the geometry initialization obtained from neural volumetric fields, and further optimize the geometry as well as a compact neural texture representation with differentiable rasterizers. Through extensive experiments, we demonstrate that our method can generate accurate mesh reconstructions with faithful appearance that are comparable to previous volume rendering methods while being an order of magnitude faster in rendering. We also show that our generated mesh and neural texture reconstruction is compatible with existing graphics pipelines and enables downstream 3D applications such as simulation. Project page: https://sarahweiii.github.io/neumanifold/
comment: Project page: https://sarahweiii.github.io/neumanifold/
♻ ☆ FireANTs: Adaptive Riemannian Optimization for Multi-Scale Diffeomorphic Matching
The paper proposes FireANTs, the first multi-scale Adaptive Riemannian Optimization algorithm for dense diffeomorphic image matching. One of the most critical and understudied aspects of diffeomorphic image matching algorithms are its highly ill-conditioned nature. We quantitatively capture the extent of ill-conditioning in a typical MRI matching task, motivating the need for an adaptive optimization algorithm for diffeomorphic matching. To this end, FireANTs generalizes the concept of momentum and adaptive estimates of the Hessian to mitigate this ill-conditioning in the non-Euclidean space of diffeomorphisms. Unlike common non-Euclidean manifolds, we also formalize considerations for multi-scale optimization of diffeomorphisms. Our rigorous mathematical results and operational contributions lead to a state-of-the-art dense matching algorithm that can be applied to generic image data with remarkable accuracy and robustness. We demonstrate consistent improvements in image matching performance across a spectrum of community-standard medical and biological correspondence matching challenges spanning a wide variety of image modalities, anatomies, resolutions, acquisition protocols, and preprocessing pipelines. This improvement is supplemented by from 300x up to 3200x speedup over existing state-of-the-art algorithms. For the first time, we perform diffeomorphic matching of sub-micron mouse cortex volumes at native resolution. Our fast implementation also enables hyperparameter studies that were intractable with existing correspondence matching algorithms.
♻ ☆ Learnable Scaled Gradient Descent for Guaranteed Robust Tensor PCA
Robust tensor principal component analysis (RTPCA) aims to separate the low-rank and sparse components from multi-dimensional data, making it an essential technique in the signal processing and computer vision fields. Recently emerging tensor singular value decomposition (t-SVD) has gained considerable attention for its ability to better capture the low-rank structure of tensors compared to traditional matrix SVD. However, existing methods often rely on the computationally expensive tensor nuclear norm (TNN), which limits their scalability for real-world tensors. To address this issue, we explore an efficient scaled gradient descent (SGD) approach within the t-SVD framework for the first time, and propose the RTPCA-SGD method. Theoretically, we rigorously establish the recovery guarantees of RTPCA-SGD under mild assumptions, demonstrating that with appropriate parameter selection, it achieves linear convergence to the true low-rank tensor at a constant rate, independent of the condition number. To enhance its practical applicability, we further propose a learnable self-supervised deep unfolding model, which enables effective parameter learning. Numerical experiments on both synthetic and real-world datasets demonstrate the superior performance of the proposed methods while maintaining competitive computational efficiency, especially consuming less time than RTPCA-TNN.
♻ ☆ Empowering Large Language Model for Continual Video Question Answering with Collaborative Prompting EMNLP 2024
In recent years, the rapid increase in online video content has underscored the limitations of static Video Question Answering (VideoQA) models trained on fixed datasets, as they struggle to adapt to new questions or tasks posed by newly available content. In this paper, we explore the novel challenge of VideoQA within a continual learning framework, and empirically identify a critical issue: fine-tuning a large language model (LLM) for a sequence of tasks often results in catastrophic forgetting. To address this, we propose Collaborative Prompting (ColPro), which integrates specific question constraint prompting, knowledge acquisition prompting, and visual temporal awareness prompting. These prompts aim to capture textual question context, visual content, and video temporal dynamics in VideoQA, a perspective underexplored in prior research. Experimental results on the NExT-QA and DramaQA datasets show that ColPro achieves superior performance compared to existing approaches, achieving 55.14\% accuracy on NExT-QA and 71.24\% accuracy on DramaQA, highlighting its practical relevance and effectiveness.
comment: Accepted by main EMNLP 2024
♻ ☆ IOR: Inversed Objects Replay for Incremental Object Detection
Existing Incremental Object Detection (IOD) methods partially alleviate catastrophic forgetting when incrementally detecting new objects in real-world scenarios. However, many of these methods rely on the assumption that unlabeled old-class objects may co-occur with labeled new-class objects in the incremental data. When unlabeled old-class objects are absent, the performance of existing methods tends to degrade. The absence can be mitigated by generating old-class samples, but it incurs high costs. This paper argues that previous generation-based IOD suffers from redundancy, both in the use of generative models, which require additional training and storage, and in the overproduction of generated samples, many of which do not contribute significantly to performance improvements. To eliminate the redundancy, we propose Inversed Objects Replay (IOR). Specifically, we generate old-class samples by inversing the original detectors, thus eliminating the necessity of training and storing additional generative models. We propose augmented replay to reuse the objects in generated samples, reducing redundant generations. Moreover, we propose high-value knowledge distillation focusing on the positions of old-class objects overwhelmed by the background, which transfers the knowledge to the incremental detector. Extensive experiments conducted on MS COCO 2017 demonstrate that our method can efficiently improve detection performance in IOD scenarios with the absence of old-class objects.
♻ ☆ Challenge Summary U-MedSAM: Uncertainty-aware MedSAM for Medical Image Segmentation
Medical Image Foundation Models have proven to be powerful tools for mask prediction across various datasets. However, accurately assessing the uncertainty of their predictions remains a significant challenge. To address this, we propose a new model, U-MedSAM, which integrates the MedSAM model with an uncertainty-aware loss function and the Sharpness-Aware Minimization (SharpMin) optimizer. The uncertainty-aware loss function automatically combines region-based, distribution-based, and pixel-based loss designs to enhance segmentation accuracy and robustness. SharpMin improves generalization by finding flat minima in the loss landscape, thereby reducing overfitting. Our method was evaluated in the CVPR24 MedSAM on Laptop challenge, where U-MedSAM demonstrated promising performance.
comment: arXiv admin note: text overlap with arXiv:2405.17496
♻ ☆ MECD+: Unlocking Event-Level Causal Graph Discovery for Video Reasoning NeurIPS 2024
Video causal reasoning aims to achieve a high-level understanding of videos from a causal perspective. However, it exhibits limitations in its scope, primarily executed in a question-answering paradigm and focusing on brief video segments containing isolated events and basic causal relations, lacking comprehensive and structured causality analysis for videos with multiple interconnected events. To fill this gap, we introduce a new task and dataset, Multi-Event Causal Discovery (MECD). It aims to uncover the causal relations between events distributed chronologically across long videos. Given visual segments and textual descriptions of events, MECD identifies the causal associations between these events to derive a comprehensive and structured event-level video causal graph explaining why and how the result event occurred. To address the challenges of MECD, we devise a novel framework inspired by the Granger Causality method, incorporating an efficient mask-based event prediction model to perform an Event Granger Test. It estimates causality by comparing the predicted result event when premise events are masked versus unmasked. Furthermore, we integrate causal inference techniques such as front-door adjustment and counterfactual inference to mitigate challenges in MECD like causality confounding and illusory causality. Additionally, context chain reasoning is introduced to conduct more robust and generalized reasoning. Experiments validate the effectiveness of our framework in reasoning complete causal relations, outperforming GPT-4o and VideoChat2 by 5.77% and 2.70%, respectively. Further experiments demonstrate that causal relation graphs can also contribute to downstream video understanding tasks such as video question answering and video event prediction.
comment: IEEE TPAMI Submission. continuous work of arXiv:2409.17647 (NeurIPS 2024)
♻ ☆ LADDER: Language Driven Slice Discovery and Error Rectification
Error slice discovery is crucial to diagnose and mitigate model errors. Current clustering or discrete attribute-based slice discovery methods face key limitations: 1) clustering results in incoherent slices, while assigning discrete attributes to slices leads to incomplete coverage of error patterns due to missing or insufficient attributes; 2) these methods lack complex reasoning, preventing them from fully explaining model biases; 3) they fail to integrate \textit{domain knowledge}, limiting their usage in specialized fields \eg radiology. We propose\ladder (\underline{La}nguage-\underline{D}riven \underline{D}iscovery and \underline{E}rror \underline{R}ectification), to address the limitations by: (1) leveraging the flexibility of natural language to address incompleteness, (2) employing LLM's latent \textit{domain knowledge} and advanced reasoning to analyze sentences and derive testable hypotheses directly, identifying biased attributes, and form coherent error slices without clustering. Existing mitigation methods typically address only the worst-performing group, often amplifying errors in other subgroups. In contrast,\ladder generates pseudo attributes from the discovered hypotheses to mitigate errors across all biases without explicit attribute annotations or prior knowledge of bias. Rigorous evaluations on 6 datasets spanning natural and medical images -- comparing 200+ classifiers with diverse architectures, pretraining strategies, and LLMs -- show that\ladder consistently outperforms existing baselines in discovering and mitigating biases.
♻ ☆ MetaNeRV: Meta Neural Representations for Videos with Spatial-Temporal Guidance AAAI2025
Neural Representations for Videos (NeRV) has emerged as a promising implicit neural representation (INR) approach for video analysis, which represents videos as neural networks with frame indexes as inputs. However, NeRV-based methods are time-consuming when adapting to a large number of diverse videos, as each video requires a separate NeRV model to be trained from scratch. In addition, NeRV-based methods spatially require generating a high-dimension signal (i.e., an entire image) from the input of a low-dimension timestamp, and a video typically consists of tens of frames temporally that have a minor change between adjacent frames. To improve the efficiency of video representation, we propose Meta Neural Representations for Videos, named MetaNeRV, a novel framework for fast NeRV representation for unseen videos. MetaNeRV leverages a meta-learning framework to learn an optimal parameter initialization, which serves as a good starting point for adapting to new videos. To address the unique spatial and temporal characteristics of video modality, we further introduce spatial-temporal guidance to improve the representation capabilities of MetaNeRV. Specifically, the spatial guidance with a multi-resolution loss aims to capture the information from different resolution stages, and the temporal guidance with an effective progressive learning strategy could gradually refine the number of fitted frames during the meta-learning process. Extensive experiments conducted on multiple datasets demonstrate the superiority of MetaNeRV for video representations and video compression.
comment: Accepted by AAAI2025
♻ ☆ Keep It Accurate and Robust: An Enhanced Nuclei Analysis Framework
Accurate segmentation and classification of nuclei in histology images is critical but challenging due to nuclei heterogeneity, staining variations, and tissue complexity. Existing methods often struggle with limited dataset variability, with patches extracted from similar whole slide images (WSI), making models prone to falling into local optima. Here we propose a new framework to address this limitation and enable robust nuclear analysis. Our method leverages dual-level ensemble modeling to overcome issues stemming from limited dataset variation. Intra-ensembling applies diverse transformations to individual samples, while inter-ensembling combines networks of different scales. We also introduce enhancements to the HoVer-Net architecture, including updated encoders, nested dense decoding and model regularization strategy. We achieve state-of-the-art results on public benchmarks, including 1st place for nuclear composition prediction and 3rd place for segmentation/classification in the 2022 Colon Nuclei Identification and Counting (CoNIC) Challenge. This success validates our approach for accurate histological nuclei analysis. Extensive experiments and ablation studies provide insights into optimal network design choices and training techniques. In conclusion, this work proposes an improved framework advancing the state-of-the-art in nuclei analysis. We release our code and models (https://github.com/WinnieLaugh/CONIC_Pathology_AI) to serve as a toolkit for the community.
Computational Geometry 4
☆ Higher Order Continuity for Smooth As-Rigid-As-Possible Shape Modeling
We propose a modification of the As-Rigid-As-Possible (ARAP) mesh deformation energy with higher order smoothness, which overcomes a prominent limitation of the original ARAP formulation: spikes and lack of continuity at the manipulation handles. Our method avoids spikes even when using single-point positional constraints. Since no explicit rotations have to be specified, the user interaction can be realized through a simple click-and-drag interface, where points on the mesh can be selected and moved around while the rest of the mesh surface automatically deforms accordingly. Our method preserves the benefits of ARAP deformations: it is easy to implement and thus useful for practical applications, while its efficiency makes it usable in real-time, interactive scenarios on detailed models.
comment: 18 pages, 11 figures
☆ A rigid origami elliptic-hyperbolic vertex duality
The field of rigid origami concerns the folding of stiff, inelastic plates of material along crease lines that act like hinges and form a straight-line planar graph, called the crease pattern of the origami. Crease pattern vertices in the interior of the folded material and that are adjacent to four crease lines, i.e. degree-4 vertices, have a single degree of freedom and can be chained together to make flexible polyhedral surfaces. Degree-4 vertices that can fold to a completely flat state have folding kinematics that are very well-understood, and thus they have been used in many engineering and physics applications. However, degree-4 vertices that are not flat-foldable or not folded from flat paper so that the vertex forms either an elliptic or hyperbolic cone, have folding angles at the creases that follow more complicated kinematic equations. In this work we present a new duality between general degree-4 rigid origami vertices, where dual vertices come in elliptic-hyperbolic pairs that have essentially equivalent kinematics. This reveals a mathematical structure in the space of degree-4 rigid origami vertices that can be leveraged in applications, for example in the construction of flexible 3D structures that possess metamaterial properties.
♻ ☆ BILTS: A Bi-Invariant Similarity Measure for Robust Object Trajectory Recognition under Reference Frame Variations
When similar object motions are performed in diverse contexts but are meant to be recognized under a single classification, these contextual variations act as disturbances that negatively affect accurate motion recognition. In this paper, we focus on contextual variations caused by reference frame variations. To robustly deal with these variations, similarity measures have been introduced that compare object motion trajectories in a context-invariant manner. However, most are highly sensitive to noise near singularities, where the measure is not uniquely defined, and lack bi-invariance (invariance to both world and body frame variations). To address these issues, we propose the novel \textit{Bi-Invariant Local Trajectory-Shape Similarity} (BILTS) measure. Compared to other measures, the BILTS measure uniquely offers bi-invariance, boundedness, and third-order shape identity. Aimed at practical implementations, we devised a discretized and regularized version of the BILTS measure which shows exceptional robustness to singularities. This is demonstrated through rigorous recognition experiments using multiple datasets. On average, BILTS attained the highest recognition ratio and least sensitivity to contextual variations compared to other invariant object motion similarity measures. We believe that the BILTS measure is a valuable tool for recognizing motions performed in diverse contexts and has potential in other applications, including the recognition, segmentation, and adaptation of both motion and force trajectories.
comment: This work has been submitted as a regular research paper for consideration in the Journal of Intelligent & Robotic Systems. The content in this preprint is identical to the version submitted for peer review, except for formatting differences required by the journal
♻ ☆ On Approximability of Steiner Tree in $\ell_p$-metrics
In the Continuous Steiner Tree problem (CST), we are given as input a set of points (called terminals) in a metric space and ask for the minimum-cost tree connecting them. Additional points (called Steiner points) from the metric space can be introduced as nodes in the solution. In the Discrete Steiner Tree problem (DST), we are given in addition to the terminals, a set of facilities, and any solution tree connecting the terminals can only contain the Steiner points from this set of facilities. Trevisan [SICOMP'00] showed that CST and DST are APX-hard when the input lies in the $\ell_1$-metric (and Hamming metric). Chleb\'ik and Chleb\'ikov\'a [TCS'08] showed that DST is NP-hard to approximate to factor of $96/95\approx 1.01$ in the graph metric (and consequently $\ell_\infty$-metric). Prior to this work, it was unclear if CST and DST are APX-hard in essentially every other popular metric. In this work, we prove that DST is APX-hard in every $\ell_p$-metric. We also prove that CST is APX-hard in the $\ell_{\infty}$-metric. Finally, we relate CST and DST, showing a general reduction from CST to DST in $\ell_p$-metrics.
comment: Abstract shortened. Journal version for TheoretiCS
Computation and Language 61
☆ BoK: Introducing Bag-of-Keywords Loss for Interpretable Dialogue Response Generation SIGDIAL 2024
The standard language modeling (LM) loss by itself has been shown to be inadequate for effective dialogue modeling. As a result, various training approaches, such as auxiliary loss functions and leveraging human feedback, are being adopted to enrich open-domain dialogue systems. One such auxiliary loss function is Bag-of-Words (BoW) loss, defined as the cross-entropy loss for predicting all the words/tokens of the next utterance. In this work, we propose a novel auxiliary loss named Bag-of-Keywords (BoK) loss to capture the central thought of the response through keyword prediction and leverage it to enhance the generation of meaningful and interpretable responses in open-domain dialogue systems. BoK loss upgrades the BoW loss by predicting only the keywords or critical words/tokens of the next utterance, intending to estimate the core idea rather than the entire response. We incorporate BoK loss in both encoder-decoder (T5) and decoder-only (DialoGPT) architecture and train the models to minimize the weighted sum of BoK and LM (BoK-LM) loss. We perform our experiments on two popular open-domain dialogue datasets, DailyDialog and Persona-Chat. We show that the inclusion of BoK loss improves the dialogue generation of backbone models while also enabling post-hoc interpretability. We also study the effectiveness of BoK-LM loss as a reference-free metric and observe comparable performance to the state-of-the-art metrics on various dialogue evaluation datasets.
comment: Accepted at SIGDIAL 2024
☆ Large language models for automated scholarly paper review: A survey
Large language models (LLMs) have significantly impacted human society, influencing various domains. Among them, academia is not simply a domain affected by LLMs, but it is also the pivotal force in the development of LLMs. In academic publications, this phenomenon is represented during the incorporation of LLMs into the peer review mechanism for reviewing manuscripts. We proposed the concept of automated scholarly paper review (ASPR) in our previous paper. As the incorporation grows, it now enters the coexistence phase of ASPR and peer review, which is described in that paper. LLMs hold transformative potential for the full-scale implementation of ASPR, but they also pose new issues and challenges that need to be addressed. In this survey paper, we aim to provide a holistic view of ASPR in the era of LLMs. We begin with a survey to find out which LLMs are used to conduct ASPR. Then, we review what ASPR-related technological bottlenecks have been solved with the incorporation of LLM technology. After that, we move on to explore new methods, new datasets, new source code, and new online systems that come with LLMs for ASPR. Furthermore, we summarize the performance and issues of LLMs in ASPR, and investigate the attitudes and reactions of publishers and academia to ASPR. Lastly, we discuss the challenges associated with the development of LLMs for ASPR. We hope this survey can serve as an inspirational reference for the researchers and promote the progress of ASPR for its actual implementation.
comment: Work in progress
☆ Hierarchical Autoregressive Transformers: Combining Byte-~and Word-Level Processing for Robust, Adaptable Language Models
Tokenization is a fundamental step in natural language processing, breaking text into units that computational models can process. While learned subword tokenizers have become the de-facto standard, they present challenges such as large vocabularies, limited adaptability to new domains or languages, and sensitivity to spelling errors and variations. To overcome these limitations, we investigate a hierarchical architecture for autoregressive language modelling that combines character-level and word-level processing. It employs a lightweight character-level encoder to convert character sequences into word embeddings, which are then processed by a word-level backbone model and decoded back into characters via a compact character-level decoder. This method retains the sequence compression benefits of word-level tokenization without relying on a rigid, predefined vocabulary. We demonstrate, at scales up to 7 billion parameters, that hierarchical transformers match the downstream task performance of subword-tokenizer-based models while exhibiting significantly greater robustness to input perturbations. Additionally, during continued pretraining on an out-of-domain language, our model trains almost twice as fast, achieves superior performance on the target language, and retains more of its previously learned knowledge. Hierarchical transformers pave the way for NLP systems that are more robust, flexible, and generalizable across languages and domains.
☆ Natural Language Processing of Privacy Policies: A Survey
Natural Language Processing (NLP) is an essential subset of artificial intelligence. It has become effective in several domains, such as healthcare, finance, and media, to identify perceptions, opinions, and misuse, among others. Privacy is no exception, and initiatives have been taken to address the challenges of usable privacy notifications to users with the help of NLP. To this aid, we conduct a literature review by analyzing 109 papers at the intersection of NLP and privacy policies. First, we provide a brief introduction to privacy policies and discuss various facets of associated problems, which necessitate the application of NLP to elevate the current state of privacy notices and disclosures to users. Subsequently, we a) provide an overview of the implementation and effectiveness of NLP approaches for better privacy policy communication; b) identify the methodologies that can be further enhanced to provide robust privacy policies; and c) identify the gaps in the current state-of-the-art research. Our systematic analysis reveals that several research papers focus on annotating and classifying privacy texts for analysis but need to adequately dwell on other aspects of NLP applications, such as summarization. More specifically, ample research opportunities exist in this domain, covering aspects such as corpus generation, summarization vectors, contextualized word embedding, identification of privacy-relevant statement categories, fine-grained classification, and domain-specific model tuning.
comment: 27 pages
☆ Towards Preventing Overreliance on Task-Oriented Conversational AI Through Accountability Modeling
Recent LLMs have enabled significant advancements for conversational agents. However, they are also well-known to hallucinate, i.e., they often produce responses that seem plausible but are not factually correct. On the other hand, users tend to over-rely on LLM-based AI agents; they accept the AI's suggestion even when it is wrong. Adding good friction, such as explanations or getting user confirmations, has been proposed as a mitigation in AI-supported decision-making systems. In this paper, we propose an accountability model for LLM-based task-oriented dialogue agents to address user overreliance via friction turns in cases of model uncertainty and errors associated with dialogue state tracking (DST). The accountability model is an augmented LLM with an additional accountability head, which functions as a binary classifier to predict the slots of the dialogue states. We perform our experiments with three backbone LLMs (Llama, Mistral, Gemma) on two established task-oriented datasets (MultiWOZ and Snips). Our empirical findings demonstrate that this approach not only enables reliable estimation of AI agent errors but also guides the LLM decoder in generating more accurate actions. We observe around 3% absolute improvement in joint goal accuracy by incorporating accountability heads in modern LLMs for the MultiWOZ dataset. We also show that this method enables the agent to self-correct its actions, further boosting its performance by 3%. Finally, we discuss the application of accountability modeling to prevent user overreliance by introducing friction.
☆ Computational Protein Science in the Era of Large Language Models (LLMs)
Considering the significance of proteins, computational protein science has always been a critical scientific field, dedicated to revealing knowledge and developing applications within the protein sequence-structure-function paradigm. In the last few decades, Artificial Intelligence (AI) has made significant impacts in computational protein science, leading to notable successes in specific protein modeling tasks. However, those previous AI models still meet limitations, such as the difficulty in comprehending the semantics of protein sequences, and the inability to generalize across a wide range of protein modeling tasks. Recently, LLMs have emerged as a milestone in AI due to their unprecedented language processing & generalization capability. They can promote comprehensive progress in fields rather than solving individual tasks. As a result, researchers have actively introduced LLM techniques in computational protein science, developing protein Language Models (pLMs) that skillfully grasp the foundational knowledge of proteins and can be effectively generalized to solve a diversity of sequence-structure-function reasoning problems. While witnessing prosperous developments, it's necessary to present a systematic overview of computational protein science empowered by LLM techniques. First, we summarize existing pLMs into categories based on their mastered protein knowledge, i.e., underlying sequence patterns, explicit structural and functional information, and external scientific languages. Second, we introduce the utilization and adaptation of pLMs, highlighting their remarkable achievements in promoting protein structure prediction, protein function prediction, and protein design studies. Then, we describe the practical application of pLMs in antibody design, enzyme design, and drug discovery. Finally, we specifically discuss the promising future directions in this fast-growing field.
☆ A Simple but Effective Closed-form Solution for Extreme Multi-label Learning ECIR25
Extreme multi-label learning (XML) is a task of assigning multiple labels from an extremely large set of labels to each data instance. Many current high-performance XML models are composed of a lot of hyperparameters, which complicates the tuning process. Additionally, the models themselves are adapted specifically to XML, which complicates their reimplementation. To remedy this problem, we propose a simple method based on ridge regression for XML. The proposed method not only has a closed-form solution but also is composed of a single hyperparameter. Since there are no precedents on applying ridge regression to XML, this paper verified the performance of the method by using various XML benchmark datasets. Furthermore, we enhanced the prediction of low-frequency labels in XML, which hold informative content. This prediction is essential yet challenging because of the limited amount of data. Here, we employed a simple frequency-based weighting. This approach greatly simplifies the process compared with existing techniques. Experimental results revealed that it can achieve levels of performance comparable to, or even exceeding, those of models with numerous hyperparameters. Additionally, we found that the frequency-based weighting significantly improved the predictive performance for low-frequency labels, while requiring almost no changes in implementation. The source code for the proposed method is available on github at https://github.com/cars1015/XML-ridge.
comment: 10pages, Accepted at ECIR25
☆ Multi-stage Training of Bilingual Islamic LLM for Neural Passage Retrieval
This study examines the use of Natural Language Processing (NLP) technology within the Islamic domain, focusing on developing an Islamic neural retrieval model. By leveraging the robust XLM-R model, the research employs a language reduction technique to create a lightweight bilingual large language model (LLM). Our approach for domain adaptation addresses the unique challenges faced in the Islamic domain, where substantial in-domain corpora exist only in Arabic while limited in other languages, including English. The work utilizes a multi-stage training process for retrieval models, incorporating large retrieval datasets, such as MS MARCO, and smaller, in-domain datasets to improve retrieval performance. Additionally, we have curated an in-domain retrieval dataset in English by employing data augmentation techniques and involving a reliable Islamic source. This approach enhances the domain-specific dataset for retrieval, leading to further performance gains. The findings suggest that combining domain adaptation and a multi-stage training method for the bilingual Islamic neural retrieval model enables it to outperform monolingual models on downstream retrieval tasks.
☆ Dual Debiasing: Remove Stereotypes and Keep Factual Gender for Fair Language Modeling and Translation
Mitigation of biases, such as language models' reliance on gender stereotypes, is a crucial endeavor required for the creation of reliable and useful language technology. The crucial aspect of debiasing is to ensure that the models preserve their versatile capabilities, including their ability to solve language tasks and equitably represent various genders. To address this issue, we introduce a streamlined Dual Dabiasing Algorithm through Model Adaptation (2DAMA). Novel Dual Debiasing enables robust reduction of stereotypical bias while preserving desired factual gender information encoded by language models. We show that 2DAMA effectively reduces gender bias in English and is one of the first approaches facilitating the mitigation of stereotypical tendencies in translation. The proposed method's key advantage is the preservation of factual gender cues, which are useful in a wide range of natural language processing tasks.
☆ ComplexFuncBench: Exploring Multi-Step and Constrained Function Calling under Long-Context Scenario
Enhancing large language models (LLMs) with real-time APIs can help generate more accurate and up-to-date responses. However, evaluating the function calling abilities of LLMs in real-world scenarios remains under-explored due to the complexity of data collection and evaluation. In this work, we introduce ComplexFuncBench, a benchmark for complex function calling across five real-world scenarios. Compared to existing benchmarks, ComplexFuncBench encompasses multi-step and constrained function calling, which requires long-parameter filing, parameter value reasoning, and 128k long context. Additionally, we propose an automatic framework, ComplexEval, for quantitatively evaluating complex function calling tasks. Through comprehensive experiments, we demonstrate the deficiencies of state-of-the-art LLMs in function calling and suggest future directions for optimizing these capabilities. The data and code are available at \url{https://github.com/THUDM/ComplexFuncBench}.
☆ BBPOS: BERT-based Part-of-Speech Tagging for Uzbek
This paper advances NLP research for the low-resource Uzbek language by evaluating two previously untested monolingual Uzbek BERT models on the part-of-speech (POS) tagging task and introducing the first publicly available UPOS-tagged benchmark dataset for Uzbek. Our fine-tuned models achieve 91% average accuracy, outperforming the baseline multi-lingual BERT as well as the rule-based tagger. Notably, these models capture intermediate POS changes through affixes and demonstrate context sensitivity, unlike existing rule-based taggers.
☆ Author-Specific Linguistic Patterns Unveiled: A Deep Learning Study on Word Class Distributions
Deep learning methods have been increasingly applied to computational linguistics to uncover patterns in text data. This study investigates author-specific word class distributions using part-of-speech (POS) tagging and bigram analysis. By leveraging deep neural networks, we classify literary authors based on POS tag vectors and bigram frequency matrices derived from their works. We employ fully connected and convolutional neural network architectures to explore the efficacy of unigram and bigram-based representations. Our results demonstrate that while unigram features achieve moderate classification accuracy, bigram-based models significantly improve performance, suggesting that sequential word class patterns are more distinctive of authorial style. Multi-dimensional scaling (MDS) visualizations reveal meaningful clustering of authors' works, supporting the hypothesis that stylistic nuances can be captured through computational methods. These findings highlight the potential of deep learning and linguistic feature analysis for author profiling and literary studies.
☆ OMoE: Diversifying Mixture of Low-Rank Adaptation by Orthogonal Finetuning
Building mixture-of-experts (MoE) architecture for Low-rank adaptation (LoRA) is emerging as a potential direction in parameter-efficient fine-tuning (PEFT) for its modular design and remarkable performance. However, simply stacking the number of experts cannot guarantee significant improvement. In this work, we first conduct qualitative analysis to indicate that experts collapse to similar representations in vanilla MoE, limiting the capacity of modular design and computational efficiency. Ulteriorly, Our analysis reveals that the performance of previous MoE variants maybe limited by a lack of diversity among experts. Motivated by these findings, we propose Orthogonal Mixture-of-Experts (OMoE), a resource-efficient MoE variant that trains experts in an orthogonal manner to promote diversity. In OMoE, a Gram-Schmidt process is leveraged to enforce that the experts' representations lie within the Stiefel manifold. By applying orthogonal constraints directly to the architecture, OMoE keeps the learning objective unchanged, without compromising optimality. Our method is simple and alleviates memory bottlenecks, as it incurs minimal experts compared to vanilla MoE models. Experiments on diverse commonsense reasoning benchmarks demonstrate that OMoE can consistently achieve stable and efficient performance improvement when compared with the state-of-the-art methods while significantly reducing the number of required experts.
☆ MSTS: A Multimodal Safety Test Suite for Vision-Language Models
Vision-language models (VLMs), which process image and text inputs, are increasingly integrated into chat assistants and other consumer AI applications. Without proper safeguards, however, VLMs may give harmful advice (e.g. how to self-harm) or encourage unsafe behaviours (e.g. to consume drugs). Despite these clear hazards, little work so far has evaluated VLM safety and the novel risks created by multimodal inputs. To address this gap, we introduce MSTS, a Multimodal Safety Test Suite for VLMs. MSTS comprises 400 test prompts across 40 fine-grained hazard categories. Each test prompt consists of a text and an image that only in combination reveal their full unsafe meaning. With MSTS, we find clear safety issues in several open VLMs. We also find some VLMs to be safe by accident, meaning that they are safe because they fail to understand even simple test prompts. We translate MSTS into ten languages, showing non-English prompts to increase the rate of unsafe model responses. We also show models to be safer when tested with text only rather than multimodal prompts. Finally, we explore the automation of VLM safety assessments, finding even the best safety classifiers to be lacking.
comment: under review
☆ Automatic Speech Recognition for Sanskrit with Transfer Learning
Sanskrit, one of humanity's most ancient languages, has a vast collection of books and manuscripts on diverse topics that have been accumulated over millennia. However, its digital content (audio and text), which is vital for the training of AI systems, is profoundly limited. Furthermore, its intricate linguistics make it hard to develop robust NLP tools for wider accessibility. Given these constraints, we have developed an automatic speech recognition model for Sanskrit by employing transfer learning mechanism on OpenAI's Whisper model. After carefully optimising the hyper-parameters, we obtained promising results with our transfer-learned model achieving a word error rate of 15.42% on Vaksancayah dataset. An online demo of our model is made available for the use of public and to evaluate its performance firsthand thereby paving the way for improved accessibility and technological support for Sanskrit learning in the modern era.
comment: Paper has been accepted at the 4th International Conference on Computer, Communication, Control & Information Technology (C3IT), Hooghly, India, 2024, pp. 1-5
☆ Attention-guided Self-reflection for Zero-shot Hallucination Detection in Large Language Models
Hallucination has emerged as a significant barrier to the effective application of Large Language Models (LLMs). In this work, we introduce a novel Attention-Guided SElf-Reflection (AGSER) approach for zero-shot hallucination detection in LLMs. The AGSER method utilizes attention contributions to categorize the input query into attentive and non-attentive queries. Each query is then processed separately through the LLMs, allowing us to compute consistency scores between the generated responses and the original answer. The difference between the two consistency scores serves as a hallucination estimator. In addition to its efficacy in detecting hallucinations, AGSER notably reduces computational complexity, requiring only three passes through the LLM and utilizing two sets of tokens. We have conducted extensive experiments with four widely-used LLMs across three different hallucination benchmarks, demonstrating that our approach significantly outperforms existing methods in zero-shot hallucination detection.
☆ Agent-as-Judge for Factual Summarization of Long Narratives
Large Language Models (LLMs) have demonstrated near-human performance in summarization tasks based on traditional metrics such as ROUGE and BERTScore. However, these metrics do not adequately capture critical aspects of summarization quality, such as factual accuracy, particularly for long narratives (>100K tokens). Recent advances, such as LLM-as-a-Judge, address the limitations of metrics based on lexical similarity but still exhibit factual inconsistencies, especially in understanding character relationships and states. In this work, we introduce NarrativeFactScore, a novel "Agent-as-a-Judge" framework for evaluating and refining summaries. By leveraging a Character Knowledge Graph (CKG) extracted from input and generated summaries, NarrativeFactScore assesses the factual consistency and provides actionable guidance for refinement, such as identifying missing or erroneous facts. We demonstrate the effectiveness of NarrativeFactScore through a detailed workflow illustration and extensive validation on widely adopted benchmarks, achieving superior performance compared to competitive methods. Our results highlight the potential of agent-driven evaluation systems to improve the factual reliability of LLM-generated summaries.
☆ RichSpace: Enriching Text-to-Video Prompt Space via Text Embedding Interpolation
Text-to-video generation models have made impressive progress, but they still struggle with generating videos with complex features. This limitation often arises from the inability of the text encoder to produce accurate embeddings, which hinders the video generation model. In this work, we propose a novel approach to overcome this challenge by selecting the optimal text embedding through interpolation in the embedding space. We demonstrate that this method enables the video generation model to produce the desired videos. Additionally, we introduce a simple algorithm using perpendicular foot embeddings and cosine similarity to identify the optimal interpolation embedding. Our findings highlight the importance of accurate text embeddings and offer a pathway for improving text-to-video generation performance.
☆ A Survey on Multi-Turn Interaction Capabilities of Large Language Models
Multi-turn interaction in the dialogue system research refers to a system's ability to maintain context across multiple dialogue turns, enabling it to generate coherent and contextually relevant responses. Recent advancements in large language models (LLMs) have significantly expanded the scope of multi-turn interaction, moving beyond chatbots to enable more dynamic agentic interactions with users or environments. In this paper, we provide a focused review of the multi-turn capabilities of LLMs, which are critical for a wide range of downstream applications, including conversational search and recommendation, consultation services, and interactive tutoring. This survey explores four key aspects: (1) the core model capabilities that contribute to effective multi-turn interaction, (2) how multi-turn interaction is evaluated in current practice, (3) the general algorithms used to enhance multi-turn interaction, and (4) potential future directions for research in this field.
comment: Draft Version, 14 pages, Ongoing refinement over time
☆ FRAG: A Flexible Modular Framework for Retrieval-Augmented Generation based on Knowledge Graphs
To mitigate the hallucination and knowledge deficiency in large language models (LLMs), Knowledge Graph (KG)-based Retrieval-Augmented Generation (RAG) has shown promising potential by utilizing KGs as external resource to enhance LLMs reasoning.However, existing KG-RAG approaches struggle with a trade-off between flexibility and retrieval quality.Modular methods prioritize flexibility by avoiding the use of KG-fine-tuned models during retrieval, leading to fixed retrieval strategies and suboptimal retrieval quality.Conversely, coupled methods embed KG information within models to improve retrieval quality, but at the expense of flexibility.In this paper, we propose a novel flexible modular KG-RAG framework, termed FRAG, which synergizes the advantages of both approaches.FRAG estimates the hop range of reasoning paths based solely on the query and classify it as either simple or complex.To match the complexity of the query, tailored pipelines are applied to ensure efficient and accurate reasoning path retrieval, thus fostering the final reasoning process.By using the query text instead of the KG to infer the structural information of reasoning paths and employing adaptable retrieval strategies, FRAG improves retrieval quality while maintaining flexibility.Moreover, FRAG does not require extra LLMs fine-tuning or calls, significantly boosting efficiency and conserving resources.Extensive experiments show that FRAG achieves state-of-the-art performance with high efficiency and low resource consumption.
☆ Sympathy over Polarization: A Computational Discourse Analysis of Social Media Posts about the July 2024 Trump Assassination Attempt
On July 13, 2024, at the Trump rally in Pennsylvania, someone attempted to assassinate Republican Presidential Candidate Donald Trump. This attempt sparked a large-scale discussion on social media. We collected posts from X (formerly known as Twitter) one week before and after the assassination attempt and aimed to model the short-term effects of such a ``shock'' on public opinions and discussion topics. Specifically, our study addresses three key questions: first, we investigate how public sentiment toward Donald Trump shifts over time and across regions (RQ1) and examine whether the assassination attempt itself significantly affects public attitudes, independent of the existing political alignments (RQ2). Finally, we explore the major themes in online conversations before and after the crisis, illustrating how discussion topics evolved in response to this politically charged event (RQ3). By integrating large language model-based sentiment analysis, difference-in-differences modeling, and topic modeling techniques, we find that following the attempt the public response was broadly sympathetic to Trump rather than polarizing, despite baseline ideological and regional disparities.
☆ Indigenous Languages Spoken in Argentina: A Survey of NLP and Speech Resources COLING
Argentina has a diverse, yet little-known, Indigenous language heritage. Most of these languages are at risk of disappearing, resulting in a significant loss of world heritage and cultural knowledge. Currently, no unified information on speakers and computational tools is available for these languages. In this work, we present a systematization of the Indigenous languages spoken in Argentina, along with national demographic data on the country's Indigenous population. The languages are classified into seven families: Mapuche, Tup\'i-Guaran\'i, Guaycur\'u, Quechua, Mataco-Mataguaya, Aymara, and Chon. We also provide an introductory survey of the computational resources available for these languages, whether or not they are specifically developed for Argentine varieties.
comment: Accepted to COLING Main 2025
☆ Passage Segmentation of Documents for Extractive Question Answering
Retrieval-Augmented Generation (RAG) has proven effective in open-domain question answering. However, the chunking process, which is essential to this pipeline, often receives insufficient attention relative to retrieval and synthesis components. This study emphasizes the critical role of chunking in improving the performance of both dense passage retrieval and the end-to-end RAG pipeline. We then introduce the Logits-Guided Multi-Granular Chunker (LGMGC), a novel framework that splits long documents into contextualized, self-contained chunks of varied granularity. Our experimental results, evaluated on two benchmark datasets, demonstrate that LGMGC not only improves the retrieval step but also outperforms existing chunking methods when integrated into a RAG pipeline.
☆ Steering Large Language Models with Feature Guided Activation Additions
Effective and reliable control over large language model (LLM) behavior is a significant challenge. While activation steering methods, which add steering vectors to a model's hidden states, are a promising approach, existing techniques often lack precision and interpretability in how they influence model outputs. We introduce Feature Guided Activation Additions (FGAA), a novel activation steering method that leverages insights from Contrastive Activation Addition (CAA) and Sparse Autoencoder-Targeted Steering (SAE-TS). By operating in the latent space of a Sparse Autoencoder (SAE) and employing optimization techniques to select desired SAE features, FGAA constructs precise steering vectors that provide better steering effects while maintaining coherence of steered model outputs. In this regard, evaluations on Gemma-2-2B and Gemma-2-9B models across various steering tasks demonstrate that FGAA outperforms existing steering methods of CAA, SAE decoder steering, and SAE-TS. Our results also highlight important trade-offs between steering scale and general model capabilities that are consistent across all tested steering methods.
comment: 7 maintext pages, 14 appendix pages
☆ Dialogue Benchmark Generation from Knowledge Graphs with Cost-Effective Retrieval-Augmented LLMs SIGMOD 2025
Dialogue benchmarks are crucial in training and evaluating chatbots engaging in domain-specific conversations. Knowledge graphs (KGs) represent semantically rich and well-organized data spanning various domains, such as DBLP, DBpedia, and YAGO. Traditionally, dialogue benchmarks have been manually created from documents, neglecting the potential of KGs in automating this process. Some question-answering benchmarks are automatically generated using extensive preprocessing from KGs, but they do not support dialogue generation. This paper introduces Chatty-Gen, a novel multi-stage retrieval-augmented generation platform for automatically generating high-quality dialogue benchmarks tailored to a specific domain using a KG. Chatty-Gen decomposes the generation process into manageable stages and uses assertion rules for automatic validation between stages. Our approach enables control over intermediate results to prevent time-consuming restarts due to hallucinations. It also reduces reliance on costly and more powerful commercial LLMs. Chatty-Gen eliminates upfront processing of the entire KG using efficient query-based retrieval to find representative subgraphs based on the dialogue context. Our experiments with several real and large KGs demonstrate that Chatty-Gen significantly outperforms state-of-the-art systems and ensures consistent model and system performance across multiple LLMs of diverse capabilities, such as GPT-4o, Gemini 1.5, Llama 3, and Mistral.
comment: The paper is publsihed in SIGMOD 2025
♻ ☆ Joint Automatic Speech Recognition And Structure Learning For Better Speech Understanding ICASSP 2025
Spoken language understanding (SLU) is a structure prediction task in the field of speech. Recently, many works on SLU that treat it as a sequence-to-sequence task have achieved great success. However, This method is not suitable for simultaneous speech recognition and understanding. In this paper, we propose a joint speech recognition and structure learning framework (JSRSL), an end-to-end SLU model based on span, which can accurately transcribe speech and extract structured content simultaneously. We conduct experiments on name entity recognition and intent classification using the Chinese dataset AISHELL-NER and the English dataset SLURP. The results show that our proposed method not only outperforms the traditional sequence-to-sequence method in both transcription and extraction capabilities but also achieves state-of-the-art performance on the two datasets.
comment: 5 pages, 2 figures, accepted by ICASSP 2025
♻ ☆ Beyond Factual Accuracy: Evaluating Coverage of Diverse Factual Information in Long-form Text Generation
This paper presents ICAT, an evaluation framework for measuring coverage of diverse factual information in long-form text generation. ICAT breaks down a long output text into a list of atomic claims and not only verifies each claim through retrieval from a (reliable) knowledge source, but also computes the alignment between the atomic factual claims and various aspects expected to be presented in the output. We study three implementations of the ICAT framework, each with a different assumption on the availability of aspects and alignment method. By adopting data from the diversification task in the TREC Web Track and the ClueWeb corpus, we evaluate the ICAT framework. We demonstrate strong correlation with human judgments and provide comprehensive evaluation across multiple state-of-the-art LLMs. Our framework further offers interpretable and fine-grained analysis of diversity and coverage. Its modular design allows for easy adaptation to different domains and datasets, making it a valuable tool for evaluating the qualitative aspects of long-form responses produced by LLMs.
♻ ☆ Neuradicon: operational representation learning of neuroimaging reports
Radiological reports typically summarize the content and interpretation of imaging studies in unstructured form that precludes quantitative analysis. This limits the monitoring of radiological services to throughput undifferentiated by content, impeding specific, targeted operational optimization. Here we present Neuradicon, a natural language processing (NLP) framework for quantitative analysis of neuroradiological reports. Our framework is a hybrid of rule-based and artificial intelligence models to represent neurological reports in succinct, quantitative form optimally suited to operational guidance. We demonstrate the application of Neuradicon to operational phenotyping of a corpus of 336,569 reports, and report excellent generalizability across time and two independent healthcare institutions.
comment: 26 pages, 11 figures
♻ ☆ Improved Paraphrase Generation via Controllable Latent Diffusion
Paraphrase generation strives to generate high-quality and diverse expressions of a given text, a domain where diffusion models excel. Though SOTA diffusion generation reconciles generation quality and diversity, textual diffusion suffers from a truncation issue that hinders efficiency and quality control. In this work, we propose \textit{L}atent \textit{D}iffusion \textit{P}araphraser~(LDP), a novel paraphrase generation by modeling a controllable diffusion process given a learned latent space. LDP achieves superior generation efficiency compared to its diffusion counterparts. It can facilitate only input segments to ensure paraphrase semantics, improving the results without external features. Experiments show that LDP better reconciles paraphrase generation quality and diversity than baselines. Further analysis shows that our method is also helpful to other similar text generations and domain adaptations
comment: The article has been accepted by Frontiers of Computer Science (FCS)
♻ ☆ Towards Large Reasoning Models: A Survey on Scaling LLM Reasoning Capabilities
Language has long been conceived as an essential tool for human reasoning. The breakthrough of Large Language Models (LLMs) has sparked significant research interest in leveraging these models to tackle complex reasoning tasks. Researchers have moved beyond simple autoregressive token generation by introducing the concept of "thought" -- a sequence of tokens representing intermediate steps in the reasoning process. This innovative paradigm enables LLMs' to mimic complex human reasoning processes, such as tree search and reflective thinking. Recently, an emerging trend of learning to reason has applied reinforcement learning (RL) to train LLMs to master reasoning processes. This approach enables the automatic generation of high-quality reasoning trajectories through trial-and-error search algorithms, significantly expanding LLMs' reasoning capacity by providing substantially more training data. Furthermore, recent studies demonstrate that encouraging LLMs to "think" with more tokens during test-time inference can further significantly boost reasoning accuracy. Therefore, the train-time and test-time scaling combined to show a new research frontier -- a path toward Large Reasoning Model. The introduction of OpenAI's o1 series marks a significant milestone in this research direction. In this survey, we present a comprehensive review of recent progress in LLM reasoning. We begin by introducing the foundational background of LLMs and then explore the key technical components driving the development of large reasoning models, with a focus on automated data construction, learning-to-reason techniques, and test-time scaling. We also analyze popular open-source projects at building large reasoning models, and conclude with open challenges and future research directions.
comment: 36 pages, 5 figures
♻ ☆ Improving Zero-Shot Chinese-English Code-Switching ASR with kNN-CTC and Gated Monolingual Datastores ICASSP 2025
The kNN-CTC model has proven to be effective for monolingual automatic speech recognition (ASR). However, its direct application to multilingual scenarios like code-switching, presents challenges. Although there is potential for performance improvement, a kNN-CTC model utilizing a single bilingual datastore can inadvertently introduce undesirable noise from the alternative language. To address this, we propose a novel kNN-CTC-based code-switching ASR (CS-ASR) framework that employs dual monolingual datastores and a gated datastore selection mechanism to reduce noise interference. Our method selects the appropriate datastore for decoding each frame, ensuring the injection of language-specific information into the ASR process. We apply this framework to cutting-edge CTC-based models, developing an advanced CS-ASR system. Extensive experiments demonstrate the remarkable effectiveness of our gated datastore mechanism in enhancing the performance of zero-shot Chinese-English CS-ASR.
comment: Accepted by ICASSP 2025
♻ ☆ Optimal Quantization for Matrix Multiplication
Recent work in machine learning community proposed multiple methods for performing lossy compression (quantization) of large matrices. This quantization is important for accelerating matrix multiplication (main component of large language models), which is often bottlenecked by the speed of loading these matrices from memory. Unlike classical vector quantization and rate-distortion theory, the goal of these new compression algorithms is to be able to approximate not the matrices themselves, but their matrix product. Specifically, given a pair of real matrices $A,B$ an encoder (compressor) is applied to each of them independently producing descriptions with $R$ bits per entry. These representations subsequently are used by the decoder to estimate matrix product $A^\top B$. In this work, we provide a non-asymptotic lower bound on the mean squared error of this approximation (as a function of rate $R$) for the case of matrices $A,B$ with iid Gaussian entries. Algorithmically, we construct a universal quantizer based on nested lattices with an explicit guarantee of approximation error for any (non-random) pair of matrices $A$, $B$ in terms of only Frobenius norms $\|\bar{A}\|_F, \|\bar{B}\|_F$ and $\|\bar{A}^\top \bar{B}\|_F$, where $\bar{A},\bar{B}$ are versions of $A,B$ with zero-centered columns, respectively. For iid Gaussian matrices our quantizer achieves the lower bound and is, thus, asymptotically optimal. A practical low-complexity version of our quantizer achieves performance quite close to optimal. In addition, we derive rate-distortion function for matrix multiplication of iid Gaussian matrices, which exhibits an interesting phase-transition at $R\approx 0.906$ bit/entry.
♻ ☆ DPCL-Diff: The Temporal Knowledge Graph Reasoning Based on Graph Node Diffusion Model with Dual-Domain Periodic Contrastive Learning
Temporal knowledge graph (TKG) reasoning that infers future missing facts is an essential and challenging task. Predicting future events typically relies on closely related historical facts, yielding more accurate results for repetitive or periodic events. However, for future events with sparse historical interactions, the effectiveness of this method, which focuses on leveraging high-frequency historical information, diminishes. Recently, the capabilities of diffusion models in image generation have opened new opportunities for TKG reasoning. Therefore, we propose a graph node diffusion model with dual-domain periodic contrastive learning (DPCL-Diff). Graph node diffusion model (GNDiff) introduces noise into sparsely related events to simulate new events, generating high-quality data that better conforms to the actual distribution. This generative mechanism significantly enhances the model's ability to reason about new events. Additionally, the dual-domain periodic contrastive learning (DPCL) maps periodic and non-periodic event entities to Poincar\'e and Euclidean spaces, leveraging their characteristics to distinguish similar periodic events effectively. Experimental results on four public datasets demonstrate that DPCL-Diff significantly outperforms state-of-the-art TKG models in event prediction, demonstrating our approach's effectiveness. This study also investigates the combined effectiveness of GNDiff and DPCL in TKG tasks.
comment: 11 pages, 2 figures
♻ ☆ Jailbreaking as a Reward Misspecification Problem
The widespread adoption of large language models (LLMs) has raised concerns about their safety and reliability, particularly regarding their vulnerability to adversarial attacks. In this paper, we propose a novel perspective that attributes this vulnerability to reward misspecification during the alignment process. This misspecification occurs when the reward function fails to accurately capture the intended behavior, leading to misaligned model outputs. We introduce a metric ReGap to quantify the extent of reward misspecification and demonstrate its effectiveness and robustness in detecting harmful backdoor prompts. Building upon these insights, we present ReMiss, a system for automated red teaming that generates adversarial prompts in a reward-misspecified space. ReMiss achieves state-of-the-art attack success rates on the AdvBench benchmark against various target aligned LLMs while preserving the human readability of the generated prompts. Furthermore, these attacks on open-source models demonstrate high transferability to closed-source models like GPT-4o and out-of-distribution tasks from HarmBench. Detailed analysis highlights the unique advantages of the proposed reward misspecification objective compared to previous methods, offering new insights for improving LLM safety and robustness.
♻ ☆ Bandit on the Hunt: Dynamic Crawling for Cyber Threat Intelligence
Public information contains valuable Cyber Threat Intelligence (CTI) that is used to prevent attacks in the future. Ideally, the learnings from previous attacks help to mitigate all those that follow. While there are standards for sharing this information, much of it is shared in non-standardized news articles or blog posts. It is a time-consuming task to monitor online sources for threats and even then, one can never be sure, to use the right sources. Current research propose extractors of Indicators of Compromise from known sources, while the identification of new sources is rarely considered. This paper proposes a focused crawler focused on the CTI domain based on multi-armed bandit ( MAB) and different crawling strategies. It uses SBERT to identify relevant documents, while dynamically adapt its crawling path. We propose a system called ThreatCrawl, which achieve a harvest rate of over 25% and is able to expand its used seed by over 300%, while retaining focus on the topic at hand. In addition, this crawler identified previously unknown but highly relevant overview pages, datasets, and domains.
comment: 6 pages, 1 figure, 3 tables
♻ ☆ How Redundant Is the Transformer Stack in Speech Representation Models? ICASSP 2025
Self-supervised speech representation models, particularly those leveraging transformer architectures, have demonstrated remarkable performance across various tasks such as speech recognition, speaker identification, and emotion detection. Recent studies on transformer models revealed a high redundancy between layers and the potential for significant pruning, which we will investigate here for transformer-based speech representation models. We perform a detailed analysis of layer similarity in speech representation models using three similarity metrics: cosine similarity, centered kernel alignment, and mutual nearest-neighbor alignment. Our findings reveal a block-like structure of high similarity, suggesting two main processing steps and significant redundancy of layers. We demonstrate the effectiveness of pruning transformer-based speech representation models without the need for post-training, achieving up to 40% reduction in transformer layers while maintaining over 95% of the model's predictive capacity. Furthermore, we employ a knowledge distillation method to substitute the entire transformer stack with mimicking layers, reducing the network size 95-98% and the inference time by up to 94%. This substantial decrease in computational load occurs without considerable performance loss, suggesting that the transformer stack is almost completely redundant for downstream applications of speech representation models.
comment: To appear at ICASSP 2025 (excluding appendix)
♻ ☆ Piece of Table: A Divide-and-Conquer Approach for Selecting Sub-Tables in Table Question Answering
Applying language models (LMs) to tables is challenging due to the inherent structural differences between two-dimensional tables and one-dimensional text for which the LMs were originally designed. Furthermore, when applying linearized tables to LMs, the maximum token lengths often imposed in self-attention calculations make it difficult to comprehensively understand the context spread across large tables. To address these challenges, we present PieTa (Piece of Table), a new framework for sub-table-based question answering (QA). PieTa operates through an iterative process of dividing tables into smaller windows, using LMs to select relevant cells within each window, and merging these cells into a sub-table. This multi-resolution approach captures dependencies across multiple rows and columns while avoiding the limitations caused by long context inputs. Instantiated as a simple iterative sub-table union algorithm, PieTa demonstrates improved performance over previous sub-table-based QA approaches.
♻ ☆ Enabling Low-Resource Language Retrieval: Establishing Baselines for Urdu MS MARCO ECIR 2025
As the Information Retrieval (IR) field increasingly recognizes the importance of inclusivity, addressing the needs of low-resource languages remains a significant challenge. This paper introduces the first large-scale Urdu IR dataset, created by translating the MS MARCO dataset through machine translation. We establish baseline results through zero-shot learning for IR in Urdu and subsequently apply the mMARCO multilingual IR methodology to this newly translated dataset. Our findings demonstrate that the fine-tuned model (Urdu-mT5-mMARCO) achieves a Mean Reciprocal Rank (MRR@10) of 0.247 and a Recall@10 of 0.439, representing significant improvements over zero-shot results and showing the potential for expanding IR access for Urdu speakers. By bridging access gaps for speakers of low-resource languages, this work not only advances multilingual IR research but also emphasizes the ethical and societal importance of inclusive IR technologies. This work provides valuable insights into the challenges and solutions for improving language representation and lays the groundwork for future research, especially in South Asian languages, which can benefit from the adaptable methods used in this study.
comment: 7 pages, ECIR 2025, conference camera-ready version
♻ ☆ VLSBench: Unveiling Visual Leakage in Multimodal Safety
Safety concerns of Multimodal large language models (MLLMs) have gradually become an important problem in various applications. Surprisingly, previous works indicate a counter-intuitive phenomenon that using textual unlearning to align MLLMs achieves comparable safety performances with MLLMs trained with image-text pairs. To explain such a counter-intuitive phenomenon, we discover a visual safety information leakage (VSIL) problem in existing multimodal safety benchmarks, i.e., the potentially risky and sensitive content in the image has been revealed in the textual query. In this way, MLLMs can easily refuse these sensitive text-image queries according to textual queries. However, image-text pairs without VSIL are common in real-world scenarios and are overlooked by existing multimodal safety benchmarks. To this end, we construct multimodal visual leakless safety benchmark (VLSBench) preventing visual safety leakage from image to textual query with 2.4k image-text pairs. Experimental results indicate that VLSBench poses a significant challenge to both open-source and close-source MLLMs, including LLaVA, Qwen2-VL, Llama3.2-Vision, and GPT-4o. This study demonstrates that textual alignment is enough for multimodal safety scenarios with VSIL, while multimodal alignment is a more promising solution for multimodal safety scenarios without VSIL. Please see our code and data at: https://hxhcreate.github.io/vlsbench.github.io/
♻ ☆ BatchLLM: Optimizing Large Batched LLM Inference with Global Prefix Sharing and Throughput-oriented Token Batching
Large language models (LLMs) increasingly play an important role in a wide range of information processing and management tasks. Many of these tasks are performed in large batches or even offline, and the performance indictor for which is throughput. These tasks usually show the characteristic of prefix sharing, where different prompt input can partially show the common prefix. However, the existing LLM inference engines tend to optimize the streaming requests and show limitations of supporting the large batched tasks with the prefix sharing characteristic. The existing solutions use the LRU-based cache to reuse the KV context of common prefix between requests. The KV context that are about to be reused may prematurely evicted with the implicit cache management. Besides, the streaming oriented systems do not leverage the request-batch information and can not mix the decoding tokens with the prefill chunks to the best for the batched scenarios, and thus fails to saturate the GPU. We propose BatchLLM to address the above problems. BatchLLM explicitly identifies the common prefixes globally. The requests sharing the same prefix will be scheduled together to reuse the KV context the best. BatchLLM reorders the requests and schedules the requests with larger ratio of decoding first to better mix the decoding tokens with the latter prefill chunks, and applies memory-centric token batching to enlarge the token-batch sizes, which helps to increase the GPU utilization. Finally, BatchLLM optimizes the prefix-shared Attention kernel with horizontal fusion to reduce tail effect and kernel launch overhead. Extensive evaluation shows that BatchLLM outperforms vLLM and SGLang by 1.3$\times$ to 10.8$\times$ on a set of microbenchmarks and a typical industry workload under different hardware environments.
♻ ☆ ELITR-Bench: A Meeting Assistant Benchmark for Long-Context Language Models COLING 2025
Research on Large Language Models (LLMs) has recently witnessed an increasing interest in extending the models' context size to better capture dependencies within long documents. While benchmarks have been proposed to assess long-range abilities, existing efforts primarily considered generic tasks that are not necessarily aligned with real-world applications. In contrast, we propose a new benchmark for long-context LLMs focused on a practical meeting assistant scenario in which the long contexts consist of transcripts obtained by automatic speech recognition, presenting unique challenges for LLMs due to the inherent noisiness and oral nature of such data. Our benchmark, ELITR-Bench, augments the existing ELITR corpus by adding 271 manually crafted questions with their ground-truth answers, as well as noisy versions of meeting transcripts altered to target different Word Error Rate levels. Our experiments with 12 long-context LLMs on ELITR-Bench confirm the progress made across successive generations of both proprietary and open models, and point out their discrepancies in terms of robustness to transcript noise. We also provide a thorough analysis of our GPT-4-based evaluation, including insights from a crowdsourcing study. Our findings indicate that while GPT-4's scores align with human judges, its ability to distinguish beyond three score levels may be limited.
comment: Published in COLING 2025
♻ ☆ Structured Packing in LLM Training Improves Long Context Utilization AAAI'25
Recent advancements in long-context large language models have attracted significant attention, yet their practical applications often suffer from suboptimal context utilization. This study investigates structuring training data to enhance semantic interdependence, demonstrating that this approach effectively improves context utilization. To this end, we introduce the Structured Packing for Long Context (SPLiCe) method, which utilizes retrieval to collate mutually relevant documents into long and coherent training examples. We validate SPLiCe empirically across models of varying sizes -- 3B, 7B, and 13B -- achieving improved performance in long-context tasks, such as Qasper and HotpotQA. Remarkably, even brief fine-tuning with SPLiCe is sufficient to realize these benefits. Additionally, SPLiCe effectively mitigates the lost-in-middle phenomenon often observed in large models. Our comprehensive analysis of SPLiCe explores its design choices and reveals intriguing transfer effects; for instance, training on programming code enhances performance on natural language tasks.
comment: AAAI'25
♻ ☆ Reinforcement Learning from Human Feedback: Whose Culture, Whose Values, Whose Perspectives?
We argue for the epistemic and ethical advantages of pluralism in Reinforcement Learning from Human Feedback (RLHF) in the context of Large Language Models (LLM). Drawing on social epistemology and pluralist philosophy of science, we suggest ways in which RHLF can be made more responsive to human needs and how we can address challenges along the way. The paper concludes with an agenda for change, i.e. concrete, actionable steps to improve LLM development.
♻ ☆ Can linguists better understand DNA?
Multilingual transfer ability, which reflects how well models fine-tuned on one source language can be applied to other languages, has been well studied in multilingual pre-trained models. However, the existence of such capability transfer between natural language and gene sequences/languages remains under explored.This study addresses this gap by drawing inspiration from the sentence-pair classification task used for evaluating sentence similarity in natural language. We constructed two analogous tasks: DNA-pair classification(DNA sequence similarity) and DNA-protein-pair classification(gene coding determination). These tasks were designed to validate the transferability of capabilities from natural language to gene sequences. Even a small-scale pre-trained model like GPT-2-small, which was pre-trained on English, achieved an accuracy of 78% on the DNA-pair classification task after being fine-tuned on English sentence-pair classification data(XTREME PAWS-X). While training a BERT model on multilingual text, the precision reached 89%. On the more complex DNA-protein-pair classification task, however, the model's output was barely distinguishable from random output.Experimental validation has confirmed that the transfer of capabilities from natural language to biological language is unequivocally present. Building on this foundation, we have also investigated the impact of model parameter scale and pre-training on this capability transfer. We provide recommendations for facilitating the transfer of capabilities from natural language to genetic language,as well as new approaches for conducting biological research based on this capability.This study offers an intriguing new perspective on exploring the relationship between natural language and genetic language.
comment: 20 pages,8 figures
♻ ☆ Text-guided Image Restoration and Semantic Enhancement for Text-to-Image Person Retrieval
The goal of Text-to-Image Person Retrieval (TIPR) is to retrieve specific person images according to the given textual descriptions. A primary challenge in this task is bridging the substantial representational gap between visual and textual modalities. The prevailing methods map texts and images into unified embedding space for matching, while the intricate semantic correspondences between texts and images are still not effectively constructed. To address this issue, we propose a novel TIPR framework to build fine-grained interactions and alignment between person images and the corresponding texts. Specifically, via fine-tuning the Contrastive Language-Image Pre-training (CLIP) model, a visual-textual dual encoder is firstly constructed, to preliminarily align the image and text features. Secondly, a Text-guided Image Restoration (TIR) auxiliary task is proposed to map abstract textual entities to specific image regions, improving the alignment between local textual and visual embeddings. Additionally, a cross-modal triplet loss is presented to handle hard samples, and further enhance the model's discriminability for minor differences. Moreover, a pruning-based text data augmentation approach is proposed to enhance focus on essential elements in descriptions, thereby avoiding excessive model attention to less significant information. The experimental results show our proposed method outperforms state-of-the-art methods on three popular benchmark datasets, and the code will be made publicly available at https://github.com/Delong-liu-bupt/SEN.
comment: The paper was withdrawn due to a dispute among the authors regarding the content of the article
♻ ☆ AceMath: Advancing Frontier Math Reasoning with Post-Training and Reward Modeling
In this paper, we introduce AceMath, a suite of frontier math models that excel in solving complex math problems, along with highly effective reward models capable of evaluating generated solutions and reliably identifying the correct ones. To develop the instruction-tuned math models, we propose a supervised fine-tuning (SFT) process that first achieves competitive performance across general domains, followed by targeted fine-tuning for the math domain using a carefully curated set of prompts and synthetically generated responses. The resulting model, AceMath-72B-Instruct greatly outperforms Qwen2.5-Math-72B-Instruct, GPT-4o and Claude-3.5 Sonnet. To develop math-specialized reward model, we first construct AceMath-RewardBench, a comprehensive and robust benchmark for evaluating math reward models across diverse problems and difficulty levels. After that, we present a systematic approach to build our math reward models. The resulting model, AceMath-72B-RM, consistently outperforms state-of-the-art reward models. Furthermore, when combining AceMath-72B-Instruct with AceMath-72B-RM, we achieve the highest average rm@8 score across the math reasoning benchmarks. We release model weights, training data, and evaluation benchmarks at: https://research.nvidia.com/labs/adlr/acemath
♻ ☆ LEGO-GraphRAG: Modularizing Graph-based Retrieval-Augmented Generation for Design Space Exploration
GraphRAG integrates (knowledge) graphs with large language models (LLMs) to improve reasoning accuracy and contextual relevance. Despite its promising applications and strong relevance to multiple research communities, such as databases and natural language processing, GraphRAG currently lacks modular workflow analysis, systematic solution frameworks, and insightful empirical studies. To bridge these gaps, we propose LEGO-GraphRAG, a modular framework that enables: 1) fine-grained decomposition of the GraphRAG workflow, 2) systematic classification of existing techniques and implemented GraphRAG instances, and 3) creation of new GraphRAG instances. Our framework facilitates comprehensive empirical studies of GraphRAG on large-scale real-world graphs and diverse query sets, revealing insights into balancing reasoning quality, runtime efficiency, and token or GPU cost, that are essential for building advanced GraphRAG systems.
♻ ☆ Empowering Large Language Model for Continual Video Question Answering with Collaborative Prompting EMNLP 2024
In recent years, the rapid increase in online video content has underscored the limitations of static Video Question Answering (VideoQA) models trained on fixed datasets, as they struggle to adapt to new questions or tasks posed by newly available content. In this paper, we explore the novel challenge of VideoQA within a continual learning framework, and empirically identify a critical issue: fine-tuning a large language model (LLM) for a sequence of tasks often results in catastrophic forgetting. To address this, we propose Collaborative Prompting (ColPro), which integrates specific question constraint prompting, knowledge acquisition prompting, and visual temporal awareness prompting. These prompts aim to capture textual question context, visual content, and video temporal dynamics in VideoQA, a perspective underexplored in prior research. Experimental results on the NExT-QA and DramaQA datasets show that ColPro achieves superior performance compared to existing approaches, achieving 55.14\% accuracy on NExT-QA and 71.24\% accuracy on DramaQA, highlighting its practical relevance and effectiveness.
comment: Accepted by main EMNLP 2024
♻ ☆ Exploring Iterative Enhancement for Improving Learnersourced Multiple-Choice Question Explanations with Large Language Models ICLR 2024
Large language models exhibit superior capabilities in processing and understanding language, yet their applications in educational contexts remain underexplored. Learnersourcing enhances learning by engaging students in creating their own educational content. When learnersourcing multiple-choice questions, creating explanations for the solution of a question is a crucial step; it helps other students understand the solution and promotes a deeper understanding of related concepts. However, it is often difficult for students to craft effective solution explanations, due to limited subject understanding. To help scaffold the task of automated explanation generation, we present and evaluate a framework called "ILearner-LLM", that iteratively enhances the generated explanations for the given questions with large language models. Comprising an explanation generation model and an explanation evaluation model, the framework generates high-quality student-aligned explanations by iteratively feeding the quality rating score from the evaluation model back into the instruction prompt of the explanation generation model. Experimental results demonstrate the effectiveness of our ILearner-LLM on LLaMA2-13B and GPT-4 to generate higher quality explanations that are closer to those written by students on five PeerWise datasets. Our findings represent a promising path to enrich the learnersourcing experience for students and to enhance the capabilities of large language models for educational applications.
comment: The short version (v4) has been accepted as a non-archival workshop paper at AGI@ICLR 2024, and the full version has been accepted by the main track of AAAI/EAAI 2025
♻ ☆ Assessing and Enhancing the Robustness of Large Language Models with Task Structure Variations for Logical Reasoning IJCAI 2023
Large language models (LLMs), such as LLaMA, Alpaca, Vicuna, GPT-3.5 and GPT-4, have advanced the performance of AI systems on various natural language processing tasks to human-like levels. However, their generalisation and robustness when performing logical reasoning has not been sufficiently assessed. To comprehensively evaluate this ability, we develop three new logical reasoning datasets named "ReClor-plus", "LogiQA-plus" and "LogiQAv2-plus" that extend standard logical reasoning datasets to evaluate the robustness of the LLM's reasoning. For each, we create three subsets: the first with randomly shuffled options, the second with the correct choices replaced by "none of the other options is correct", and the third with a combination of shuffling and substitution. Experiments on these datasets show that these simple augmentations greatly hinder the models' performance. Despite their high performance on the original publicly available datasets, we find that all models perform poorly on these newly constructed datasets. We also demonstrate that introducing task variations into the training set can markedly improve the model's performance on both the original and our developed datasets. Finally, we show that applying logic-driven data augmentation for fine-tuning and prompting can enhance generalisation in both discriminative and generative models, offering a path to improving their robustness for tasks involving logical reasoning. Source code and data are made publicly available at https://github.com/Strong-AI-Lab/Logical-and-abstract-reasoning.
comment: The short version (v3) was accepted for oral presentation at the first LLM@IJCAI 2023 non-archival symposium, and the full version was accepted by ICONIP 2024
♻ ☆ GRASP: A Grid-Based Benchmark for Evaluating Commonsense Spatial Reasoning
Spatial reasoning, an important faculty of human cognition with many practical applications, is one of the core commonsense skills that is not purely language-based and, for satisfying (as opposed to optimal) solutions, requires some minimum degree of planning. Existing benchmarks of Commonsense Spatial Reasoning (CSR) tend to evaluate how Large Language Models (LLMs) interpret text-based spatial $\textit{descriptions}$ rather than directly evaluate a plan produced by the LLM in response to a $\textit{specific}$ spatial reasoning problem. In this paper, we construct a large-scale benchmark called GRASP, which consists of 16,000 grid-based environments where the agent is tasked with an energy collection problem. These environments include 100 grid instances instantiated using each of the 160 different grid settings, involving five different energy distributions, two modes of agent starting position, and two distinct obstacle configurations, as well as three kinds of agent constraints. Using GRASP, we compare classic baseline approaches, such as random walk and greedy search methods, with advanced LLMs like GPT-3.5-Turbo, GPT-4o, and GPT-o1-mini. The experimental results indicate that even these advanced LLMs struggle to consistently achieve satisfactory solutions.
♻ ☆ T3: A Novel Zero-shot Transfer Learning Framework Iteratively Training on an Assistant Task for a Target Task
Long text summarization, gradually being essential for efficiently processing large volumes of information, stays challenging for Large Language Models (LLMs) such as GPT and LLaMA families because of the insufficient open-sourced training datasets and the high requirement of contextual details dealing. To address the issue, we design a novel zero-shot transfer learning framework, abbreviated as T3, to iteratively training a baseline LLM on an assistant task for the target task, where the former should own richer data resources and share structural or semantic similarity with the latter. In practice, T3 is approached to deal with the long text summarization task by utilizing question answering as the assistant task, and further validated its effectiveness on the BBC summary, NarraSum, FairytaleQA, and NLQuAD datasets, with up to nearly 14% improvement in ROUGE, 35% improvement in BLEU, and 16% improvement in Factscore compared to three baseline LLMs, demonstrating its potential for more assistant-target task combinations.
♻ ☆ Can AI-Generated Text be Reliably Detected?
Large Language Models (LLMs) perform impressively well in various applications. However, the potential for misuse of these models in activities such as plagiarism, generating fake news, and spamming has raised concern about their responsible use. Consequently, the reliable detection of AI-generated text has become a critical area of research. AI text detectors have shown to be effective under their specific settings. In this paper, we stress-test the robustness of these AI text detectors in the presence of an attacker. We introduce recursive paraphrasing attack to stress test a wide range of detection schemes, including the ones using the watermarking as well as neural network-based detectors, zero shot classifiers, and retrieval-based detectors. Our experiments conducted on passages, each approximately 300 tokens long, reveal the varying sensitivities of these detectors to our attacks. Our findings indicate that while our recursive paraphrasing method can significantly reduce detection rates, it only slightly degrades text quality in many cases, highlighting potential vulnerabilities in current detection systems in the presence of an attacker. Additionally, we investigate the susceptibility of watermarked LLMs to spoofing attacks aimed at misclassifying human-written text as AI-generated. We demonstrate that an attacker can infer hidden AI text signatures without white-box access to the detection method, potentially leading to reputational risks for LLM developers. Finally, we provide a theoretical framework connecting the AUROC of the best possible detector to the Total Variation distance between human and AI text distributions. This analysis offers insights into the fundamental challenges of reliable detection as language models continue to advance. Our code is publicly available at https://github.com/vinusankars/Reliability-of-AI-text-detectors.
comment: Published in Transactions on Machine Learning Research (TMLR)
♻ ☆ Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluators
Large Language Models (LLMs) have demonstrated promising capabilities as automatic evaluators in assessing the quality of generated natural language. However, LLMs still exhibit biases in evaluation and often struggle to generate coherent evaluations that align with human assessments. In this work, we first conduct a systematic study of the misalignment between LLM evaluators and human evaluation, revealing that existing calibration methods aimed at mitigating biases of LLMs are insufficient for effectively aligning LLM evaluators. Inspired by the use of preference data in RLHF, we formulate the evaluation as a ranking problem and introduce Pairwise-preference Search (PAIRS), an uncertainty-guided search-based rank aggregation method that employs LLMs to conduct pairwise comparisons locally and efficiently ranks candidate texts globally. PAIRS achieves state-of-the-art performance on representative evaluation tasks in long-form generations and demonstrates significant improvements over direct scoring. Furthermore, we provide insights into the role of pairwise preference in quantifying the transitivity of LLMs and demonstrate how PAIRS benefits from calibration using debiased pairwise evaluations.
comment: This paper has been accepted by COLM 2024
♻ ☆ NL2KQL: From Natural Language to Kusto Query
Data is growing rapidly in volume and complexity. Proficiency in database query languages is pivotal for crafting effective queries. As coding assistants become more prevalent, there is significant opportunity to enhance database query languages. The Kusto Query Language (KQL) is a widely used query language for large semi-structured data such as logs, telemetries, and time-series for big data analytics platforms. This paper introduces NL2KQL an innovative framework that uses large language models (LLMs) to convert natural language queries (NLQs) to KQL queries. The proposed NL2KQL framework includes several key components: Schema Refiner which narrows down the schema to its most pertinent elements; the Few-shot Selector which dynamically selects relevant examples from a few-shot dataset; and the Query Refiner which repairs syntactic and semantic errors in KQL queries. Additionally, this study outlines a method for generating large datasets of synthetic NLQ-KQL pairs which are valid within a specific database contexts. To validate NL2KQL's performance, we utilize an array of online (based on query execution) and offline (based on query parsing) metrics. Through ablation studies, the significance of each framework component is examined, and the datasets used for benchmarking are made publicly available. This work is the first of its kind and is compared with available baselines to demonstrate its effectiveness.
♻ ☆ Fast Matrix Multiplications for Lookup Table-Quantized LLMs EMNLP 2024
The deployment of large language models (LLMs) is often constrained by memory bandwidth, where the primary bottleneck is the cost of transferring model parameters from the GPU's global memory to its registers. When coupled with custom kernels that fuse the dequantization and matmul operations, weight-only quantization can thus enable faster inference by reducing the amount of memory movement. However, developing high-performance kernels for weight-quantized LLMs presents substantial challenges, especially when the weights are compressed to non-evenly-divisible bit widths (e.g., 3 bits) with non-uniform, lookup table (LUT) quantization. This paper describes FLUTE, a flexible lookup table engine for LUT-quantized LLMs, which uses offline restructuring of the quantized weight matrix to minimize bit manipulations associated with unpacking, and vectorization and duplication of the lookup table to mitigate shared memory bandwidth constraints. At batch sizes < 32 and quantization group size of 128 (typical in LLM inference), the FLUTE kernel can be 2-4x faster than existing GEMM kernels. As an application of FLUTE, we explore a simple extension to lookup table-based NormalFloat quantization and apply it to quantize LLaMA3 to various configurations, obtaining competitive quantization performance against strong baselines while obtaining an end-to-end throughput increase of 1.5 to 2 times.
comment: EMNLP 2024 (Findings)
♻ ☆ Analyzing Continuous Semantic Shifts with Diachronic Word Similarity Matrices COLING2025
The meanings and relationships of words shift over time. This phenomenon is referred to as semantic shift. Research focused on understanding how semantic shifts occur over multiple time periods is essential for gaining a detailed understanding of semantic shifts. However, detecting change points only between adjacent time periods is insufficient for analyzing detailed semantic shifts, and using BERT-based methods to examine word sense proportions incurs a high computational cost. To address those issues, we propose a simple yet intuitive framework for how semantic shifts occur over multiple time periods by leveraging a similarity matrix between the embeddings of the same word through time. We compute a diachronic word similarity matrix using fast and lightweight word embeddings across arbitrary time periods, making it deeper to analyze continuous semantic shifts. Additionally, by clustering the similarity matrices for different words, we can categorize words that exhibit similar behavior of semantic shift in an unsupervised manner.
comment: COLING2025
♻ ☆ LADDER: Language Driven Slice Discovery and Error Rectification
Error slice discovery is crucial to diagnose and mitigate model errors. Current clustering or discrete attribute-based slice discovery methods face key limitations: 1) clustering results in incoherent slices, while assigning discrete attributes to slices leads to incomplete coverage of error patterns due to missing or insufficient attributes; 2) these methods lack complex reasoning, preventing them from fully explaining model biases; 3) they fail to integrate \textit{domain knowledge}, limiting their usage in specialized fields \eg radiology. We propose\ladder (\underline{La}nguage-\underline{D}riven \underline{D}iscovery and \underline{E}rror \underline{R}ectification), to address the limitations by: (1) leveraging the flexibility of natural language to address incompleteness, (2) employing LLM's latent \textit{domain knowledge} and advanced reasoning to analyze sentences and derive testable hypotheses directly, identifying biased attributes, and form coherent error slices without clustering. Existing mitigation methods typically address only the worst-performing group, often amplifying errors in other subgroups. In contrast,\ladder generates pseudo attributes from the discovered hypotheses to mitigate errors across all biases without explicit attribute annotations or prior knowledge of bias. Rigorous evaluations on 6 datasets spanning natural and medical images -- comparing 200+ classifiers with diverse architectures, pretraining strategies, and LLMs -- show that\ladder consistently outperforms existing baselines in discovering and mitigating biases.
♻ ☆ LLM Hallucinations in Practical Code Generation: Phenomena, Mechanism, and Mitigation ISSTA 2025
Code generation aims to automatically generate code from input requirements, significantly enhancing development efficiency. Recent large language models (LLMs) based approaches have shown promising results and revolutionized code generation task. Despite the promising performance, LLMs often generate contents with hallucinations, especially for the code generation scenario requiring the handling of complex contextual dependencies in practical development process. Although previous study has analyzed hallucinations in LLM-powered code generation, the study is limited to standalone function generation. In this paper, we conduct an empirical study to study the phenomena, mechanism, and mitigation of LLM hallucinations within more practical and complex development contexts in repository-level generation scenario. First, we manually examine the code generation results from six mainstream LLMs to establish a hallucination taxonomy of LLM-generated code. Next, we elaborate on the phenomenon of hallucinations, analyze their distribution across different models. We then analyze causes of hallucinations and identify four potential factors contributing to hallucinations. Finally, we propose an RAG-based mitigation method, which demonstrates consistent effectiveness in all studied LLMs. The replication package including code, data, and experimental results is available at https://github.com/DeepSoftwareAnalytics/LLMCodingHallucination
comment: Accepted by ISSTA 2025
♻ ☆ Keeping LLMs Aligned After Fine-tuning: The Crucial Role of Prompt Templates NeurIPS 2024
Public LLMs such as the Llama 2-Chat underwent alignment training and were considered safe. Recently Qi et al. [2024] reported that even benign fine-tuning on seemingly safe datasets can give rise to unsafe behaviors in the models. The current paper is about methods and best practices to mitigate such loss of alignment. We focus on the setting where a public model is fine-tuned before serving users for specific usage, where the model should improve on the downstream task while maintaining alignment. Through extensive experiments on several chat models (Meta's Llama 2-Chat, Mistral AI's Mistral 7B Instruct v0.2, and OpenAI's GPT-3.5 Turbo), this paper uncovers that the prompt templates used during fine-tuning and inference play a crucial role in preserving safety alignment, and proposes the ``Pure Tuning, Safe Testing'' (PTST) strategy -- fine-tune models without a safety prompt, but include it at test time. This seemingly counterintuitive strategy incorporates an intended distribution shift to encourage alignment preservation. Fine-tuning experiments on GSM8K, ChatDoctor, and OpenOrca show that PTST significantly reduces the rise of unsafe behaviors.
comment: NeurIPS 2024
♻ ☆ RLPF: Reinforcement Learning from Prediction Feedback for User Summarization with LLMs AAAI 2025
LLM-powered personalization agent systems employ Large Language Models (LLMs) to predict users' behavior from their past activities. However, their effectiveness often hinges on the ability to effectively leverage extensive, long user historical data due to its inherent noise and length of such data. Existing pretrained LLMs may generate summaries that are concise but lack the necessary context for downstream tasks, hindering their utility in personalization systems. To address these challenges, we introduce Reinforcement Learning from Prediction Feedback (RLPF). RLPF fine-tunes LLMs to generate concise, human-readable user summaries that are optimized for downstream task performance. By maximizing the usefulness of the generated summaries, RLPF effectively distills extensive user history data while preserving essential information for downstream tasks. Our empirical evaluation demonstrates significant improvements in both extrinsic downstream task utility and intrinsic summary quality, surpassing baseline methods by up to 22% on downstream task performance and achieving an up to 84.59% win rate on Factuality, Abstractiveness, and Readability. RLPF also achieves a remarkable 74% reduction in context length while improving performance on 16 out of 19 unseen tasks and/or datasets, showcasing its generalizability. This approach offers a promising solution for enhancing LLM personalization by effectively transforming long, noisy user histories into informative and human-readable representations.
comment: AAAI 2025
Machine Learning 118
☆ Credit Risk Identification in Supply Chains Using Generative Adversarial Networks
Credit risk management within supply chains has emerged as a critical research area due to its significant implications for operational stability and financial sustainability. The intricate interdependencies among supply chain participants mean that credit risks can propagate across networks, with impacts varying by industry. This study explores the application of Generative Adversarial Networks (GANs) to enhance credit risk identification in supply chains. GANs enable the generation of synthetic credit risk scenarios, addressing challenges related to data scarcity and imbalanced datasets. By leveraging GAN-generated data, the model improves predictive accuracy while effectively capturing dynamic and temporal dependencies in supply chain data. The research focuses on three representative industries-manufacturing (steel), distribution (pharmaceuticals), and services (e-commerce) to assess industry-specific credit risk contagion. Experimental results demonstrate that the GAN-based model outperforms traditional methods, including logistic regression, decision trees, and neural networks, achieving superior accuracy, recall, and F1 scores. The findings underscore the potential of GANs in proactive risk management, offering robust tools for mitigating financial disruptions in supply chains. Future research could expand the model by incorporating external market factors and supplier relationships to further enhance predictive capabilities. Keywords- Generative Adversarial Networks (GANs); Supply Chain Risk; Credit Risk Identification; Machine Learning; Data Augmentation
comment: The paper will be published and indexed by IEEE at 2025 8th International Conference on Advanced Algorithms and Control Engineering (ICAACE 2025)
☆ ColNet: Collaborative Optimization in Decentralized Federated Multi-task Learning Systems
The integration of Federated Learning (FL) and Multi-Task Learning (MTL) has been explored to address client heterogeneity, with Federated Multi-Task Learning (FMTL) treating each client as a distinct task. However, most existing research focuses on data heterogeneity (e.g., addressing non-IID data) rather than task heterogeneity, where clients solve fundamentally different tasks. Additionally, much of the work relies on centralized settings with a server managing the federation, leaving the more challenging domain of decentralized FMTL largely unexplored. Thus, this work bridges this gap by proposing ColNet, a framework designed for heterogeneous tasks in decentralized federated environments. ColNet divides models into the backbone and task-specific layers, forming groups of similar clients, with group leaders performing conflict-averse cross-group aggregation. A pool of experiments with different federations demonstrated ColNet outperforms the compared aggregation schemes in decentralized settings with label and task heterogeneity scenarios.
☆ Hybrid Deep Learning Model for epileptic seizure classification by using 1D-CNN with multi-head attention mechanism
Epilepsy is a prevalent neurological disorder globally, impacting around 50 million people \cite{WHO_epilepsy_50million}. Epileptic seizures result from sudden abnormal electrical activity in the brain, which can be read as sudden and significant changes in the EEG signal of the brain. The signal can vary in severity and frequency, which results in loss of consciousness and muscle contractions for a short period of time \cite{epilepsyfoundation_myoclonic}. Individuals with epilepsy often face significant employment challenges due to safety concerns in certain work environments. Many jobs that involve working at heights, operating heavy machinery, or in other potentially hazardous settings may be restricted for people with seizure disorders. This certainly limits job options and economic opportunities for those living with epilepsy.
☆ New Fashion Products Performance Forecasting: A Survey on Evolutions, Models and Emerging Trends
The fast fashion industry's insatiable demand for new styles and rapid production cycles has led to a significant environmental burden. Overproduction, excessive waste, and harmful chemicals have contributed to the negative environmental impact of the industry. To mitigate these issues, a paradigm shift that prioritizes sustainability and efficiency is urgently needed. Integrating learning-based predictive analytics into the fashion industry represents a significant opportunity to address environmental challenges and drive sustainable practices. By forecasting fashion trends and optimizing production, brands can reduce their ecological footprint while remaining competitive in a rapidly changing market. However, one of the key challenges in forecasting fashion sales is the dynamic nature of consumer preferences. Fashion is acyclical, with trends constantly evolving and resurfacing. In addition, cultural changes and unexpected events can disrupt established patterns. This problem is also known as New Fashion Products Performance Forecasting (NFPPF), and it has recently gained more and more interest in the global research landscape. Given its multidisciplinary nature, the field of NFPPF has been approached from many different angles. This comprehensive survey wishes to provide an up-to-date overview that focuses on learning-based NFPPF strategies. The survey is based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodological flow, allowing for a systematic and complete literature review. In particular, we propose the first taxonomy that covers the learning panorama for NFPPF, examining in detail the different methodologies used to increase the amount of multimodal information, as well as the state-of-the-art available datasets. Finally, we discuss the challenges and future directions.
comment: Accepted at the Springer Nature Computer Science journal
☆ Hierarchical Autoregressive Transformers: Combining Byte-~and Word-Level Processing for Robust, Adaptable Language Models
Tokenization is a fundamental step in natural language processing, breaking text into units that computational models can process. While learned subword tokenizers have become the de-facto standard, they present challenges such as large vocabularies, limited adaptability to new domains or languages, and sensitivity to spelling errors and variations. To overcome these limitations, we investigate a hierarchical architecture for autoregressive language modelling that combines character-level and word-level processing. It employs a lightweight character-level encoder to convert character sequences into word embeddings, which are then processed by a word-level backbone model and decoded back into characters via a compact character-level decoder. This method retains the sequence compression benefits of word-level tokenization without relying on a rigid, predefined vocabulary. We demonstrate, at scales up to 7 billion parameters, that hierarchical transformers match the downstream task performance of subword-tokenizer-based models while exhibiting significantly greater robustness to input perturbations. Additionally, during continued pretraining on an out-of-domain language, our model trains almost twice as fast, achieves superior performance on the target language, and retains more of its previously learned knowledge. Hierarchical transformers pave the way for NLP systems that are more robust, flexible, and generalizable across languages and domains.
☆ Towards Human-Guided, Data-Centric LLM Co-Pilots
Machine learning (ML) has the potential to revolutionize healthcare, but its adoption is often hindered by the disconnect between the needs of domain experts and translating these needs into robust and valid ML tools. Despite recent advances in LLM-based co-pilots to democratize ML for non-technical domain experts, these systems remain predominantly focused on model-centric aspects while overlooking critical data-centric challenges. This limitation is problematic in complex real-world settings where raw data often contains complex issues, such as missing values, label noise, and domain-specific nuances requiring tailored handling. To address this we introduce CliMB-DC, a human-guided, data-centric framework for LLM co-pilots that combines advanced data-centric tools with LLM-driven reasoning to enable robust, context-aware data processing. At its core, CliMB-DC introduces a novel, multi-agent reasoning system that combines a strategic coordinator for dynamic planning and adaptation with a specialized worker agent for precise execution. Domain expertise is then systematically incorporated to guide the reasoning process using a human-in-the-loop approach. To guide development, we formalize a taxonomy of key data-centric challenges that co-pilots must address. Thereafter, to address the dimensions of the taxonomy, we integrate state-of-the-art data-centric tools into an extensible, open-source architecture, facilitating the addition of new tools from the research community. Empirically, using real-world healthcare datasets we demonstrate CliMB-DC's ability to transform uncurated datasets into ML-ready formats, significantly outperforming existing co-pilot baselines for handling data-centric challenges. CliMB-DC promises to empower domain experts from diverse domains -- healthcare, finance, social sciences and more -- to actively participate in driving real-world impact using ML.
comment: Saveliev, Liu & Seedat contributed equally
☆ Pairwise Elimination with Instance-Dependent Guarantees for Bandits with Cost Subsidy
Multi-armed bandits (MAB) are commonly used in sequential online decision-making when the reward of each decision is an unknown random variable. In practice however, the typical goal of maximizing total reward may be less important than minimizing the total cost of the decisions taken, subject to a reward constraint. For example, we may seek to make decisions that have at least the reward of a reference ``default'' decision, with as low a cost as possible. This problem was recently introduced in the Multi-Armed Bandits with Cost Subsidy (MAB-CS) framework. MAB-CS is broadly applicable to problem domains where a primary metric (cost) is constrained by a secondary metric (reward), and the rewards are unknown. In our work, we address variants of MAB-CS including ones with reward constrained by the reward of a known reference arm or by the subsidized best reward. We introduce the Pairwise-Elimination (PE) algorithm for the known reference arm variant and generalize PE to PE-CS for the subsidized best reward variant. Our instance-dependent analysis of PE and PE-CS reveals that both algorithms have an order-wise logarithmic upper bound on Cost and Quality Regret, making our policies the first with such a guarantee. Moreover, by comparing our upper and lower bound results we establish that PE is order-optimal for all known reference arm problem instances. Finally, experiments are conducted using the MovieLens 25M and Goodreads datasets for both PE and PE-CS revealing the effectiveness of PE and the superior balance between performance and reliability offered by PE-CS compared to baselines from the literature.
☆ SEANN: A Domain-Informed Neural Network for Epidemiological Insights
In epidemiology, traditional statistical methods such as logistic regression, linear regression, and other parametric models are commonly employed to investigate associations between predictors and health outcomes. However, non-parametric machine learning techniques, such as deep neural networks (DNNs), coupled with explainable AI (XAI) tools, offer new opportunities for this task. Despite their potential, these methods face challenges due to the limited availability of high-quality, high-quantity data in this field. To address these challenges, we introduce SEANN, a novel approach for informed DNNs that leverages a prevalent form of domain-specific knowledge: Pooled Effect Sizes (PES). PESs are commonly found in published Meta-Analysis studies, in different forms, and represent a quantitative form of a scientific consensus. By direct integration within the learning procedure using a custom loss, we experimentally demonstrate significant improvements in the generalizability of predictive performances and the scientific plausibility of extracted relationships compared to a domain-knowledge agnostic neural network in a scarce and noisy data setting.
☆ Logarithmic Regret for Nonlinear Control
We address the problem of learning to control an unknown nonlinear dynamical system through sequential interactions. Motivated by high-stakes applications in which mistakes can be catastrophic, such as robotics and healthcare, we study situations where it is possible for fast sequential learning to occur. Fast sequential learning is characterized by the ability of the learning agent to incur logarithmic regret relative to a fully-informed baseline. We demonstrate that fast sequential learning is achievable in a diverse class of continuous control problems where the system dynamics depend smoothly on unknown parameters, provided the optimal control policy is persistently exciting. Additionally, we derive a regret bound which grows with the square root of the number of interactions for cases where the optimal policy is not persistently exciting. Our results provide the first regret bounds for controlling nonlinear dynamical systems depending nonlinearly on unknown parameters. We validate the trends our theory predicts in simulation on a simple dynamical system.
☆ DADA: Dual Averaging with Distance Adaptation
We present a novel universal gradient method for solving convex optimization problems. Our algorithm -- Dual Averaging with Distance Adaptation (DADA) -- is based on the classical scheme of dual averaging and dynamically adjusts its coefficients based on observed gradients and the distance between iterates and the starting point, eliminating the need for problem-specific parameters. DADA is a universal algorithm that simultaneously works for a broad spectrum of problem classes, provided the local growth of the objective function around its minimizer can be bounded. Particular examples of such problem classes are nonsmooth Lipschitz functions, Lipschitz-smooth functions, H\"older-smooth functions, functions with high-order Lipschitz derivative, quasi-self-concordant functions, and $(L_0,L_1)$-smooth functions. Crucially, DADA is applicable to both unconstrained and constrained problems, even when the domain is unbounded, without requiring prior knowledge of the number of iterations or desired accuracy.
☆ Unsupervised Rhythm and Voice Conversion of Dysarthric to Healthy Speech for ASR ICASSP 2025
Automatic speech recognition (ASR) systems are well known to perform poorly on dysarthric speech. Previous works have addressed this by speaking rate modification to reduce the mismatch with typical speech. Unfortunately, these approaches rely on transcribed speech data to estimate speaking rates and phoneme durations, which might not be available for unseen speakers. Therefore, we combine unsupervised rhythm and voice conversion methods based on self-supervised speech representations to map dysarthric to typical speech. We evaluate the outputs with a large ASR model pre-trained on healthy speech without further fine-tuning and find that the proposed rhythm conversion especially improves performance for speakers of the Torgo corpus with more severe cases of dysarthria. Code and audio samples are available at https://idiap.github.io/RnV .
comment: Accepted at ICASSP 2025 Satellite Workshop: Workshop on Speech Pathology Analysis and DEtection (SPADE)
☆ Over-the-Air Multi-Sensor Inference with Neural Networks Using Memristor-Based Analog Computing
Deep neural networks provide reliable solutions for many classification and regression tasks; however, their application in real-time wireless systems with simple sensor networks is limited due to high energy consumption and significant bandwidth needs. This study proposes a multi-sensor wireless inference system with memristor-based analog computing. Given the sensors' limited computational capabilities, the features from the network's front end are transmitted to a central device where an $L_p$-norm inspired approximation of the maximum operation is employed to achieve transformation-invariant features, enabling efficient over-the-air transmission. We also introduce a trainable over-the-air sensor fusion method based on $L_p$-norm inspired combining function that customizes sensor fusion to match the network and sensor distribution characteristics, enhancing adaptability. To address the energy constraints of sensors, we utilize memristors, known for their energy-efficient in-memory computing, enabling analog-domain computations that reduce energy use and computational overhead in edge computing. This dual approach of memristors and $L_p$-norm inspired sensor fusion fosters energy-efficient computational and transmission paradigms and serves as a practical energy-efficient solution with minimal performance loss.
comment: 34 pages
☆ Challenges and recommendations for Electronic Health Records data extraction and preparation for dynamic prediction modelling in hospitalized patients -- a practical guide
Dynamic predictive modeling using electronic health record (EHR) data has gained significant attention in recent years. The reliability and trustworthiness of such models depend heavily on the quality of the underlying data, which is largely determined by the stages preceding the model development: data extraction from EHR systems and data preparation. We list over forty challenges encountered during these stages and provide actionable recommendations for addressing them. These challenges are organized into four categories: cohort definition, outcome definition, feature engineering, and data cleaning. This list is designed to serve as a practical guide for data extraction engineers and researchers, supporting better practices and improving the quality and real-world applicability of dynamic prediction models in clinical settings.
☆ SpaceTime: Causal Discovery from Non-Stationary Time Series
Understanding causality is challenging and often complicated by changing causal relationships over time and across environments. Climate patterns, for example, shift over time with recurring seasonal trends, while also depending on geographical characteristics such as ecosystem variability. Existing methods for discovering causal graphs from time series either assume stationarity, do not permit both temporal and spatial distribution changes, or are unaware of locations with the same causal relationships. In this work, we therefore unify the three tasks of causal graph discovery in the non-stationary multi-context setting, of reconstructing temporal regimes, and of partitioning datasets and time intervals into those where invariant causal relationships hold. To construct a consistent score that forms the basis of our method, we employ the Minimum Description Length principle. Our resulting algorithm SPACETIME simultaneously accounts for heterogeneity across space and non-stationarity over time. Given multiple time series, it discovers regime changepoints and a temporal causal graph using non-parametric functional modeling and kernelized discrepancy testing. We also show that our method provides insights into real-world phenomena such as river-runoff measured at different catchments and biosphere-atmosphere interactions across ecosystems.
☆ Counterfactual Explanations for k-means and Gaussian Clustering
Counterfactuals have been recognized as an effective approach to explain classifier decisions. Nevertheless, they have not yet been considered in the context of clustering. In this work, we propose the use of counterfactuals to explain clustering solutions. First, we present a general definition for counterfactuals for model-based clustering that includes plausibility and feasibility constraints. Then we consider the counterfactual generation problem for k-means and Gaussian clustering assuming Euclidean distance. Our approach takes as input the factual, the target cluster, a binary mask indicating actionable or immutable features and a plausibility factor specifying how far from the cluster boundary the counterfactual should be placed. In the k-means clustering case, analytical mathematical formulas are presented for computing the optimal solution, while in the Gaussian clustering case (assuming full, diagonal, or spherical covariances) our method requires the numerical solution of a nonlinear equation with a single parameter only. We demonstrate the advantages of our approach through illustrative examples and quantitative experimental comparisons.
☆ Amortized Bayesian Mixture Models
Finite mixtures are a broad class of models useful in scenarios where observed data is generated by multiple distinct processes but without explicit information about the responsible process for each data point. Estimating Bayesian mixture models is computationally challenging due to issues such as high-dimensional posterior inference and label switching. Furthermore, traditional methods such as MCMC are applicable only if the likelihoods for each mixture component are analytically tractable. Amortized Bayesian Inference (ABI) is a simulation-based framework for estimating Bayesian models using generative neural networks. This allows the fitting of models without explicit likelihoods, and provides fast inference. ABI is therefore an attractive framework for estimating mixture models. This paper introduces a novel extension of ABI tailored to mixture models. We factorize the posterior into a distribution of the parameters and a distribution of (categorical) mixture indicators, which allows us to use a combination of generative neural networks for parameter inference, and classification networks for mixture membership identification. The proposed framework accommodates both independent and dependent mixture models, enabling filtering and smoothing. We validate and demonstrate our approach through synthetic and real-world datasets.
comment: 34 pages, 17 figures
☆ Modelling Activity Scheduling Behaviour with Deep Generative Machine Learning
We model human activity scheduling behaviour using a deep generative machine learning approach. Activity schedules, which represent the activities and associated travel behaviours of individuals, are a core component of many applied models in the transport, energy and epidemiology domains. Our data driven approach learns human preferences and scheduling logic without the need for complex interacting combinations of sub-models and custom-rules, this makes our approach significantly faster and simpler to operate that existing approaches. We find activity schedule data combines aspects of both continuous image data and also discrete text data, requiring novel approaches. We additionally contribute a novel schedule representation and comprehensive evaluation framework for generated schedules. Evaluation shows our approach is able to rapidly generate large, diverse and realistic synthetic samples of activity schedules.
☆ The Relevance of AWS Chronos: An Evaluation of Standard Methods for Time Series Forecasting with Limited Tuning
A systematic comparison of Chronos, a transformer-based time series forecasting framework, against traditional approaches including ARIMA and Prophet. We evaluate these models across multiple time horizons and user categories, with a focus on the impact of historical context length. Our analysis reveals that while Chronos demonstrates superior performance for longer-term predictions and maintains accuracy with increased context, traditional models show significant degradation as context length increases. We find that prediction quality varies systematically between user classes, suggesting that underlying behavior patterns always influence model performance. This study provides a case for deploying Chronos in real-world applications where limited model tuning is feasible, especially in scenarios requiring longer prediction.
☆ Temporal Graph MLP Mixer for Spatio-Temporal Forecasting
Spatiotemporal forecasting is critical in applications such as traffic prediction, climate modeling, and environmental monitoring. However, the prevalence of missing data in real-world sensor networks significantly complicates this task. In this paper, we introduce the Temporal Graph MLP-Mixer (T-GMM), a novel architecture designed to address these challenges. The model combines node-level processing with patch-level subgraph encoding to capture localized spatial dependencies while leveraging a three-dimensional MLP-Mixer to handle temporal, spatial, and feature-based dependencies. Experiments on the AQI, ENGRAD, PV-US and METR-LA datasets demonstrate the model's ability to effectively forecast even in the presence of significant missing data. While not surpassing state-of-the-art models in all scenarios, the T-GMM exhibits strong learning capabilities, particularly in capturing long-range dependencies. These results highlight its potential for robust, scalable spatiotemporal forecasting.
☆ Hypercone Assisted Contour Generation for Out-of-Distribution Detection
Recent advances in the field of out-of-distribution (OOD) detection have placed great emphasis on learning better representations suited to this task. While there are distance-based approaches, distributional awareness has seldom been exploited for better performance. We present HAC$_k$-OOD, a novel OOD detection method that makes no distributional assumption about the data, but automatically adapts to its distribution. Specifically, HAC$_k$-OOD constructs a set of hypercones by maximizing the angular distance to neighbors in a given data-point's vicinity to approximate the contour within which in-distribution (ID) data-points lie. Experimental results show state-of-the-art FPR@95 and AUROC performance on Near-OOD detection and on Far-OOD detection on the challenging CIFAR-100 benchmark without explicitly training for OOD performance.
☆ Provably Safeguarding a Classifier from OOD and Adversarial Samples: an Extreme Value Theory Approach
This paper introduces a novel method, Sample-efficient Probabilistic Detection using Extreme Value Theory (SPADE), which transforms a classifier into an abstaining classifier, offering provable protection against out-of-distribution and adversarial samples. The approach is based on a Generalized Extreme Value (GEV) model of the training distribution in the classifier's latent space, enabling the formal characterization of OOD samples. Interestingly, under mild assumptions, the GEV model also allows for formally characterizing adversarial samples. The abstaining classifier, which rejects samples based on their assessment by the GEV model, provably avoids OOD and adversarial samples. The empirical validation of the approach, conducted on various neural architectures (ResNet, VGG, and Vision Transformer) and medium and large-sized datasets (CIFAR-10, CIFAR-100, and ImageNet), demonstrates its frugality, stability, and efficiency compared to the state of the art.
comment: under review
☆ Contributions to the Decision Theoretic Foundations of Machine Learning and Robust Statistics under Weakly Structured Information
This habilitation thesis is cumulative and, therefore, is collecting and connecting research that I (together with several co-authors) have conducted over the last few years. Thus, the absolute core of the work is formed by the ten publications listed on page 5 under the name Contributions 1 to 10. The references to the complete versions of these articles are also found in this list, making them as easily accessible as possible for readers wishing to dive deep into the different research projects. The chapters following this thesis, namely Parts A to C and the concluding remarks, serve to place the articles in a larger scientific context, to (briefly) explain their respective content on a less formal level, and to highlight some interesting perspectives for future research in their respective contexts. Naturally, therefore, the following presentation has neither the level of detail nor the formal rigor that can (hopefully) be found in the papers. The purpose of the following text is to provide the reader an easy and high-level access to this interesting and important research field as a whole, thereby, advertising it to a broader audience.
comment: Habilitation Thesis
☆ Surrogate-based multiscale analysis of experiments on thermoplastic composites under off-axis loading
In this paper, we present a surrogate-based multiscale approach to model constant strain-rate and creep experiments on unidirectional thermoplastic composites under off-axis loading. In previous contributions, these experiments were modeled through a single-scale micromechanical simulation under the assumption of macroscopic homogeneity. Although efficient and accurate in many scenarios, simulations with low-off axis angles showed significant discrepancies with the experiments. It was hypothesized that the mismatch was caused by macroscopic inhomogeneity, which would require a multiscale approach to capture it. However, full-field multiscale simulations remain computationally prohibitive. To address this issue, we replace the micromodel with a Physically Recurrent Neural Network (PRNN), a surrogate model that combines data-driven components with embedded constitutive models to capture history-dependent behavior naturally. The explainability of the latent space of this network is also explored in a transfer learning strategy that requires no re-training. With the surrogate-based simulations, we confirm the hypothesis raised on the inhomogeneity of the macroscopic strain field and gain insights into the influence of adjustment of the experimental setup with oblique end-tabs. Results from the surrogate-based multiscale approach show better agreement with experiments than the single-scale micromechanical approach over a wide range of settings, although with limited accuracy on the creep experiments, where macroscopic test effects were implicitly taken into account in the material properties calibration.
comment: 21 pages. 31 figures
☆ Improved learning rates in multi-unit uniform price auctions NeurIPS 2024
Motivated by the strategic participation of electricity producers in electricity day-ahead market, we study the problem of online learning in repeated multi-unit uniform price auctions focusing on the adversarial opposing bid setting. The main contribution of this paper is the introduction of a new modeling of the bid space. Indeed, we prove that a learning algorithm leveraging the structure of this problem achieves a regret of $\tilde{O}(K^{4/3}T^{2/3})$ under bandit feedback, improving over the bound of $\tilde{O}(K^{7/4}T^{3/4})$ previously obtained in the literature. This improved regret rate is tight up to logarithmic terms. Inspired by electricity reserve markets, we further introduce a different feedback model under which all winning bids are revealed. This feedback interpolates between the full-information and bandit scenarios depending on the auctions' results. We prove that, under this feedback, the algorithm that we propose achieves regret $\tilde{O}(K^{5/2}\sqrt{T})$.
comment: NeurIPS 2024
☆ A Simple but Effective Closed-form Solution for Extreme Multi-label Learning ECIR25
Extreme multi-label learning (XML) is a task of assigning multiple labels from an extremely large set of labels to each data instance. Many current high-performance XML models are composed of a lot of hyperparameters, which complicates the tuning process. Additionally, the models themselves are adapted specifically to XML, which complicates their reimplementation. To remedy this problem, we propose a simple method based on ridge regression for XML. The proposed method not only has a closed-form solution but also is composed of a single hyperparameter. Since there are no precedents on applying ridge regression to XML, this paper verified the performance of the method by using various XML benchmark datasets. Furthermore, we enhanced the prediction of low-frequency labels in XML, which hold informative content. This prediction is essential yet challenging because of the limited amount of data. Here, we employed a simple frequency-based weighting. This approach greatly simplifies the process compared with existing techniques. Experimental results revealed that it can achieve levels of performance comparable to, or even exceeding, those of models with numerous hyperparameters. Additionally, we found that the frequency-based weighting significantly improved the predictive performance for low-frequency labels, while requiring almost no changes in implementation. The source code for the proposed method is available on github at https://github.com/cars1015/XML-ridge.
comment: 10pages, Accepted at ECIR25
☆ Mean and Variance Estimation Complexity in Arbitrary Distributions via Wasserstein Minimization
Parameter estimation is a fundamental challenge in machine learning, crucial for tasks such as neural network weight fitting and Bayesian inference. This paper focuses on the complexity of estimating translation $\boldsymbol{\mu} \in \mathbb{R}^l$ and shrinkage $\sigma \in \mathbb{R}_{++}$ parameters for a distribution of the form $\frac{1}{\sigma^l} f_0 \left( \frac{\boldsymbol{x} - \boldsymbol{\mu}}{\sigma} \right)$, where $f_0$ is a known density in $\mathbb{R}^l$ given $n$ samples. We highlight that while the problem is NP-hard for Maximum Likelihood Estimation (MLE), it is possible to obtain $\varepsilon$-approximations for arbitrary $\varepsilon > 0$ within $\text{poly} \left( \frac{1}{\varepsilon} \right)$ time using the Wasserstein distance.
☆ Convex Physics Informed Neural Networks for the Monge-Ampère Optimal Transport Problem
Optimal transportation of raw material from suppliers to customers is an issue arising in logistics that is addressed here with a continuous model relying on optimal transport theory. A physics informed neuralnetwork method is advocated here for the solution of the corresponding generalized Monge-Amp`ere equation. Convex neural networks are advocated to enforce the convexity of the solution to the Monge-Amp\`ere equation and obtain a suitable approximation of the optimal transport map. A particular focus is set on the enforcement of transport boundary conditions in the loss function. Numerical experiments illustrate the solution to the optimal transport problem in several configurations, and sensitivity analyses are performed.
comment: 17 pages, 14 figures. Submitted to Engineering Computations on 26 September 2024
☆ Region-wise stacking ensembles for estimating brain-age using MRI
Predictive modeling using structural magnetic resonance imaging (MRI) data is a prominent approach to study brain-aging. Machine learning algorithms and feature extraction methods have been employed to improve predictions and explore healthy and accelerated aging e.g. neurodegenerative and psychiatric disorders. The high-dimensional MRI data pose challenges to building generalizable and interpretable models as well as for data privacy. Common practices are resampling or averaging voxels within predefined parcels, which reduces anatomical specificity and biological interpretability as voxels within a region may differently relate to aging. Effectively, naive fusion by averaging can result in information loss and reduced accuracy. We present a conceptually novel two-level stacking ensemble (SE) approach. The first level comprises regional models for predicting individuals' age based on voxel-wise information, fused by a second-level model yielding final predictions. Eight data fusion scenarios were explored using as input Gray matter volume (GMV) estimates from four datasets covering the adult lifespan. Performance, measured using mean absolute error (MAE), R2, correlation and prediction bias, showed that SE outperformed the region-wise averages. The best performance was obtained when first-level regional predictions were obtained as out-of-sample predictions on the application site with second-level models trained on independent and site-specific data (MAE=4.75 vs baseline regional mean GMV MAE=5.68). Performance improved as more datasets were used for training. First-level predictions showed improved and more robust aging signal providing new biological insights and enhanced data privacy. Overall, the SE improves accuracy compared to the baseline while preserving or enhancing data privacy.
comment: version1
☆ Enhancing UAV Path Planning Efficiency Through Accelerated Learning
Unmanned Aerial Vehicles (UAVs) are increasingly essential in various fields such as surveillance, reconnaissance, and telecommunications. This study aims to develop a learning algorithm for the path planning of UAV wireless communication relays, which can reduce storage requirements and accelerate Deep Reinforcement Learning (DRL) convergence. Assuming the system possesses terrain maps of the area and can estimate user locations using localization algorithms or direct GPS reporting, it can input these parameters into the learning algorithms to achieve optimized path planning performance. However, higher resolution terrain maps are necessary to extract topological information such as terrain height, object distances, and signal blockages. This requirement increases memory and storage demands on UAVs while also lengthening convergence times in DRL algorithms. Similarly, defining the telecommunication coverage map in UAV wireless communication relays using these terrain maps and user position estimations demands higher memory and storage utilization for the learning path planning algorithms. Our approach reduces path planning training time by applying a dimensionality reduction technique based on Principal Component Analysis (PCA), sample combination, Prioritized Experience Replay (PER), and the combination of Mean Squared Error (MSE) and Mean Absolute Error (MAE) loss calculations in the coverage map estimates, thereby enhancing a Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm. The proposed solution reduces the convergence episodes needed for basic training by approximately four times compared to the traditional TD3.
comment: This paper was accepted in https://camad2024.ieee-camad.org/ conference but it is not available from the conference yet
☆ Conformal Prediction Sets with Improved Conditional Coverage using Trust Scores
Standard conformal prediction offers a marginal guarantee on coverage, but for prediction sets to be truly useful, they should ideally ensure coverage conditional on each test point. Unfortunately, it is impossible to achieve exact, distribution-free conditional coverage in finite samples. In this work, we propose an alternative conformal prediction algorithm that targets coverage where it matters most--in instances where a classifier is overconfident in its incorrect predictions. We start by dissecting miscoverage events in marginally-valid conformal prediction, and show that miscoverage rates vary based on the classifier's confidence and its deviation from the Bayes optimal classifier. Motivated by this insight, we develop a variant of conformal prediction that targets coverage conditional on a reduced set of two variables: the classifier's confidence in a prediction and a nonparametric trust score that measures its deviation from the Bayes classifier. Empirical evaluation on multiple image datasets shows that our method generally improves conditional coverage properties compared to standard conformal prediction, including class-conditional coverage, coverage over arbitrary subgroups, and coverage over demographic groups.
☆ Visual Exploration of Stopword Probabilities in Topic Models
Stopword removal is a critical stage in many Machine Learning methods but often receives little consideration, it interferes with the model visualizations and disrupts user confidence. Inappropriately chosen or hastily omitted stopwords not only lead to suboptimal performance but also significantly affect the quality of models, thus reducing the willingness of practitioners and stakeholders to rely on the output visualizations. This paper proposes a novel extraction method that provides a corpus-specific probabilistic estimation of stopword likelihood and an interactive visualization system to support their analysis. We evaluated our approach and interface using real-world data, a commonly used Machine Learning method (Topic Modelling), and a comprehensive qualitative experiment probing user confidence. The results of our work show that our system increases user confidence in the credibility of topic models by (1) returning reasonable probabilities, (2) generating an appropriate and representative extension of common stopword lists, and (3) providing an adjustable threshold for estimating and analyzing stopwords visually. Finally, we discuss insights, recommendations, and best practices to support practitioners while improving the output of Machine Learning methods and topic model visualizations with robust stopword analysis and removal.
☆ Exploring the Impact of Generative Artificial Intelligence in Education: A Thematic Analysis
The recent advancements in Generative Artificial intelligence (GenAI) technology have been transformative for the field of education. Large Language Models (LLMs) such as ChatGPT and Bard can be leveraged to automate boilerplate tasks, create content for personalised teaching, and handle repetitive tasks to allow more time for creative thinking. However, it is important to develop guidelines, policies, and assessment methods in the education sector to ensure the responsible integration of these tools. In this article, thematic analysis has been performed on seven essays obtained from professionals in the education sector to understand the advantages and pitfalls of using GenAI models such as ChatGPT and Bard in education. Exploratory Data Analysis (EDA) has been performed on the essays to extract further insights from the text. The study found several themes which highlight benefits and drawbacks of GenAI tools, as well as suggestions to overcome these limitations and ensure that students are using these tools in a responsible and ethical manner.
☆ Gene Regulatory Network Inference in the Presence of Selection Bias and Latent Confounders
Gene Regulatory Network Inference (GRNI) aims to identify causal relationships among genes using gene expression data, providing insights into regulatory mechanisms. A significant yet often overlooked challenge is selection bias, a process where only cells meeting specific criteria, such as gene expression thresholds, survive or are observed, distorting the true joint distribution of genes and thus biasing GRNI results. Furthermore, gene expression is influenced by latent confounders, such as non-coding RNAs, which add complexity to GRNI. To address these challenges, we propose GISL (Gene Regulatory Network Inference in the presence of Selection bias and Latent confounders), a novel algorithm to infer true regulatory relationships in the presence of selection and confounding issues. Leveraging data obtained via multiple gene perturbation experiments, we show that the true regulatory relationships, as well as selection processes and latent confounders can be partially identified without strong parametric models and under mild graphical assumptions. Experimental results on both synthetic and real-world single-cell gene expression datasets demonstrate the superiority of GISL over existing methods.
☆ PaSa: An LLM Agent for Comprehensive Academic Paper Search
We introduce PaSa, an advanced Paper Search agent powered by large language models. PaSa can autonomously make a series of decisions, including invoking search tools, reading papers, and selecting relevant references, to ultimately obtain comprehensive and accurate results for complex scholarly queries. We optimize PaSa using reinforcement learning with a synthetic dataset, AutoScholarQuery, which includes 35k fine-grained academic queries and corresponding papers sourced from top-tier AI conference publications. Additionally, we develop RealScholarQuery, a benchmark collecting real-world academic queries to assess PaSa performance in more realistic scenarios. Despite being trained on synthetic data, PaSa significantly outperforms existing baselines on RealScholarQuery, including Google, Google Scholar, Google with GPT-4 for paraphrased queries, chatGPT (search-enabled GPT-4o), GPT-o1, and PaSa-GPT-4o (PaSa implemented by prompting GPT-4o). Notably, PaSa-7B surpasses the best Google-based baseline, Google with GPT-4o, by 37.78% in recall@20 and 39.90% in recall@50. It also exceeds PaSa-GPT-4o by 30.36% in recall and 4.25% in precision. Model, datasets, and code are available at https://github.com/bytedance/pasa.
☆ Robotic World Model: A Neural Network Simulator for Robust Policy Optimization in Robotics
Learning robust and generalizable world models is crucial for enabling efficient and scalable robotic control in real-world environments. In this work, we introduce a novel framework for learning world models that accurately capture complex, partially observable, and stochastic dynamics. The proposed method employs a dual-autoregressive mechanism and self-supervised training to achieve reliable long-horizon predictions without relying on domain-specific inductive biases, ensuring adaptability across diverse robotic tasks. We further propose a policy optimization framework that leverages world models for efficient training in imagined environments and seamless deployment in real-world systems. Through extensive experiments, our approach consistently outperforms state-of-the-art methods, demonstrating superior autoregressive prediction accuracy, robustness to noise, and generalization across manipulation and locomotion tasks. Notably, policies trained with our method are successfully deployed on ANYmal D hardware in a zero-shot transfer, achieving robust performance with minimal sim-to-real performance loss. This work advances model-based reinforcement learning by addressing the challenges of long-horizon prediction, error accumulation, and sim-to-real transfer. By providing a scalable and robust framework, the introduced methods pave the way for adaptive and efficient robotic systems in real-world applications.
☆ landmarker: a Toolkit for Anatomical Landmark Localization in 2D/3D Images
Anatomical landmark localization in 2D/3D images is a critical task in medical imaging. Although many general-purpose tools exist for landmark localization in classical computer vision tasks, such as pose estimation, they lack the specialized features and modularity necessary for anatomical landmark localization applications in the medical domain. Therefore, we introduce landmarker, a Python package built on PyTorch. The package provides a comprehensive, flexible toolkit for developing and evaluating landmark localization algorithms, supporting a range of methodologies, including static and adaptive heatmap regression. landmarker enhances the accuracy of landmark identification, streamlines research and development processes, and supports various image formats and preprocessing pipelines. Its modular design allows users to customize and extend the toolkit for specific datasets and applications, accelerating innovation in medical imaging. landmarker addresses a critical need for precision and customization in landmark localization tasks not adequately met by existing general-purpose pose estimation tools.
comment: 11 pages, 4 figures
☆ A recursive Bayesian neural network for constitutive modeling of sands under monotonic loading
In geotechnical engineering, constitutive models play a crucial role in describing soil behavior under varying loading conditions. Data-driven deep learning (DL) models offer a promising alternative for developing predictive constitutive models. When prediction is the primary focus, quantifying the predictive uncertainty of a trained DL model and communicating this uncertainty to end users is crucial for informed decision-making. This study proposes a recursive Bayesian neural network (rBNN) framework, which builds upon recursive feedforward neural networks (rFFNNs) by introducing generalized Bayesian inference for uncertainty quantification. A significant contribution of this work is the incorporation of a sliding window approach in rFFNNs, allowing the models to effectively capture temporal dependencies across load steps. The rBNN extends this framework by treating model parameters as random variables, with their posterior distributions inferred using generalized variational inference. The proposed framework is validated on two datasets: (i) a numerically simulated consolidated drained (CD) triaxial dataset employing a hardening soil model and (ii) an experimental dataset comprising 28 CD triaxial tests on Baskarp sand. Comparative analyses with LSTM, Bi-LSTM, and GRU models demonstrate that the deterministic rFFNN achieves superior predictive accuracy, attributed to its transparent structure and sliding window design. While the rBNN marginally trails in accuracy for the experimental case, it provides robust confidence intervals, addressing data sparsity and measurement noise in experimental conditions. The study underscores the trade-offs between deterministic and probabilistic approaches and the potential of rBNNs for uncertainty-aware constitutive modeling.
☆ Two-level Solar Irradiance Clustering with Season Identification: A Comparative Analysis
Solar irradiance clustering can enhance solar power capacity planning and help improve forecasting models by identifying similar irradiance patterns influenced by seasonal and weather changes. In this study, we adopt an efficient two-level clustering approach to automatically identify seasons using the clear sky irradiance in first level and subsequently to identify daily cloud level as clear, cloudy and partly cloudy within each season in second level. In the second level of clustering, three methods are compared, namely, Daily Irradiance Index (DII or $\beta$), Euclidean Distance (ED), and Dynamic Time Warping (DTW) distance. The DII is computed as the ratio of time integral of measured irradiance to time integral of the clear sky irradiance. The identified clusters were compared quantitatively using established clustering metrics and qualitatively by comparing the mean irradiance profiles. The results clearly establish the superiority of the $\beta$-based clustering approach as the leader, setting a new benchmark for solar irradiance clustering studies. Moreover, $\beta$-based clustering remains effective even for annual data unlike the time-series methods which suffer significant performance degradation. Interestingly, contrary to expectations, ED-based clustering outperforms the more compute-intensive DTW distance-based clustering. The method has been rigorously validated using data from two distinct US locations, demonstrating robust scalability for larger datasets and potential applicability for other locations.
comment: 30 pages, 9 figures, 6 tables
☆ Double descent in quantum machine learning
The double descent phenomenon challenges traditional statistical learning theory by revealing scenarios where larger models do not necessarily lead to reduced performance on unseen data. While this counterintuitive behavior has been observed in a variety of classical machine learning models, particularly modern neural network architectures, it remains elusive within the context of quantum machine learning. In this work, we analytically demonstrate that quantum learning models can exhibit double descent behavior by drawing on insights from linear regression and random matrix theory. Additionally, our numerical experiments on quantum kernel methods across different real-world datasets and system sizes further confirm the existence of a test error peak, a characteristic feature of double descent. Our findings provide evidence that quantum models can operate in the modern, overparameterized regime without experiencing overfitting, thereby opening pathways to improved learning performance beyond traditional statistical learning theory.
☆ Robust Change Captioning in Remote Sensing: SECOND-CC Dataset and MModalCC Framework
Remote sensing change captioning (RSICC) aims to describe changes between bitemporal images in natural language. Existing methods often fail under challenges like illumination differences, viewpoint changes, blur effects, leading to inaccuracies, especially in no-change regions. Moreover, the images acquired at different spatial resolutions and have registration errors tend to affect the captions. To address these issues, we introduce SECOND-CC, a novel RSICC dataset featuring high-resolution RGB image pairs, semantic segmentation maps, and diverse real-world scenarios. SECOND-CC which contains 6,041 pairs of bitemporal RS images and 30,205 sentences describing the differences between images. Additionally, we propose MModalCC, a multimodal framework that integrates semantic and visual data using advanced attention mechanisms, including Cross-Modal Cross Attention (CMCA) and Multimodal Gated Cross Attention (MGCA). Detailed ablation studies and attention visualizations further demonstrate its effectiveness and ability to address RSICC challenges. Comprehensive experiments show that MModalCC outperforms state-of-the-art RSICC methods, including RSICCformer, Chg2Cap, and PSNet with +4.6% improvement on BLEU4 score and +9.6% improvement on CIDEr score. We will make our dataset and codebase publicly available to facilitate future research at https://github.com/ChangeCapsInRS/SecondCC
comment: This work has been submitted to the IEEE Transactions on Geoscience and Remote Sensing journal for possible publication
☆ One-D-Piece: Image Tokenizer Meets Quality-Controllable Compression
Current image tokenization methods require a large number of tokens to capture the information contained within images. Although the amount of information varies across images, most image tokenizers only support fixed-length tokenization, leading to inefficiency in token allocation. In this study, we introduce One-D-Piece, a discrete image tokenizer designed for variable-length tokenization, achieving quality-controllable mechanism. To enable variable compression rate, we introduce a simple but effective regularization mechanism named "Tail Token Drop" into discrete one-dimensional image tokenizers. This method encourages critical information to concentrate at the head of the token sequence, enabling support of variadic tokenization, while preserving state-of-the-art reconstruction quality. We evaluate our tokenizer across multiple reconstruction quality metrics and find that it delivers significantly better perceptual quality than existing quality-controllable compression methods, including JPEG and WebP, at smaller byte sizes. Furthermore, we assess our tokenizer on various downstream computer vision tasks, including image classification, object detection, semantic segmentation, and depth estimation, confirming its adaptability to numerous applications compared to other variable-rate methods. Our approach demonstrates the versatility of variable-length discrete image tokenization, establishing a new paradigm in both compression efficiency and reconstruction performance. Finally, we validate the effectiveness of tail token drop via detailed analysis of tokenizers.
comment: Our Project Page: https://turingmotors.github.io/one-d-piece-tokenizer
☆ OMoE: Diversifying Mixture of Low-Rank Adaptation by Orthogonal Finetuning
Building mixture-of-experts (MoE) architecture for Low-rank adaptation (LoRA) is emerging as a potential direction in parameter-efficient fine-tuning (PEFT) for its modular design and remarkable performance. However, simply stacking the number of experts cannot guarantee significant improvement. In this work, we first conduct qualitative analysis to indicate that experts collapse to similar representations in vanilla MoE, limiting the capacity of modular design and computational efficiency. Ulteriorly, Our analysis reveals that the performance of previous MoE variants maybe limited by a lack of diversity among experts. Motivated by these findings, we propose Orthogonal Mixture-of-Experts (OMoE), a resource-efficient MoE variant that trains experts in an orthogonal manner to promote diversity. In OMoE, a Gram-Schmidt process is leveraged to enforce that the experts' representations lie within the Stiefel manifold. By applying orthogonal constraints directly to the architecture, OMoE keeps the learning objective unchanged, without compromising optimality. Our method is simple and alleviates memory bottlenecks, as it incurs minimal experts compared to vanilla MoE models. Experiments on diverse commonsense reasoning benchmarks demonstrate that OMoE can consistently achieve stable and efficient performance improvement when compared with the state-of-the-art methods while significantly reducing the number of required experts.
☆ Accelerating Large Language Models through Partially Linear Feed-Forward Network
Large language models (LLMs) demonstrate remarkable capabilities but face deployment challenges due to their massive parameter counts. While existing compression techniques like pruning can reduce model size, it leads to significant accuracy degradation under high compression ratios. We present a novel perspective inspired by constant folding in compiler optimization. Our approach enables parameter reduction by treating activation functions in LLMs as linear functions. However, recent LLMs use complex non-linear activations like GELU that prevent direct application of this technique. We propose TARDIS, which enables optimization of LLMs with non-linear activations by partially approximating them with linear functions in frequently occurring input ranges. For outlier inputs, TARDIS employs an online predictor to dynamically fall back to original computations. Our experiments demonstrate that TARDIS achieves 80% parameter reduction in feed-forward networks, while significantly outperforming state-of-the-art pruning methods Wanda and RIA with up to 65% higher accuracy. In practical deployments for a 7B model, TARDIS achieves 1.6x end-to-end inference speedup when integrated with the vLLM serving system, and 1.4x speedup with the widely adopted HuggingFace implementation, while incurring only a 10.9% accuracy trade-off.
☆ Tracking student skills real-time through a continuous-variable dynamic Bayesian network
The field of Knowledge Tracing is focused on predicting the success rate of a student for a given skill. Modern methods like Deep Knowledge Tracing provide accurate estimates given enough data, but being based on neural networks they struggle to explain how these estimates are formed. More classical methods like Dynamic Bayesian Networks can do this, but they cannot give data on the accuracy of their estimates and often struggle to incorporate new observations in real-time due to their high computational load. This paper presents a novel method, Performance Distribution Tracing (PDT), in which the distribution of the success rate is traced live. It uses a Dynamic Bayesian Network with continuous random variables as nodes. By tracing the success rate distribution, there is always data available on the accuracy of any success rate estimation. In addition, it makes it possible to combine data from similar/related skills to come up with a more informed estimate of success rates. This makes it possible to predict exercise success rates, providing both explainability and an accuracy indication, even when an exercise requires a combination of different skills to solve. And through the use of the beta distribution functions as conjugate priors, all distributions are available in analytical form, allowing efficient online updates upon new observations. Experiments have shown that the resulting estimates generally feel sufficiently accurate to end-users such that they accept recommendations based on them.
☆ PandaSkill -- Player Performance and Skill Rating in Esports: Application to League of Legends
To take the esports scene to the next level, we introduce PandaSkill, a framework for assessing player performance and skill rating. Traditional rating systems like Elo and TrueSkill often overlook individual contributions and face challenges in professional esports due to limited game data and fragmented competitive scenes. PandaSkill leverages machine learning to estimate in-game player performance from individual player statistics. Each in-game role is modeled independently, ensuring a fair comparison between them. Then, using these performance scores, PandaSkill updates the player skill ratings using the Bayesian framework OpenSkill in a free-for-all setting. In this setting, skill ratings are updated solely based on performance scores rather than game outcomes, hightlighting individual contributions. To address the challenge of isolated rating pools that hinder cross-regional comparisons, PandaSkill introduces a dual-rating system that combines players' regional ratings with a meta-rating representing each region's overall skill level. Applying PandaSkill to five years of professional League of Legends matches worldwide, we show that our method produces skill ratings that better predict game outcomes and align more closely with expert opinions compared to existing methods.
☆ Virtual Nodes Improve Long-term Traffic Prediction
Effective traffic prediction is a cornerstone of intelligent transportation systems, enabling precise forecasts of traffic flow, speed, and congestion. While traditional spatio-temporal graph neural networks (ST-GNNs) have achieved notable success in short-term traffic forecasting, their performance in long-term predictions remains limited. This challenge arises from over-squashing problem, where bottlenecks and limited receptive fields restrict information flow and hinder the modeling of global dependencies. To address these challenges, this study introduces a novel framework that incorporates virtual nodes, which are additional nodes added to the graph and connected to existing nodes, in order to aggregate information across the entire graph within a single GNN layer. Our proposed model incorporates virtual nodes by constructing a semi-adaptive adjacency matrix. This matrix integrates distance-based and adaptive adjacency matrices, allowing the model to leverage geographical information while also learning task-specific features from data. Experimental results demonstrate that the inclusion of virtual nodes significantly enhances long-term prediction accuracy while also improving layer-wise sensitivity to mitigate the over-squashing problem. Virtual nodes also offer enhanced explainability by focusing on key intersections and high-traffic areas, as shown by the visualization of their adjacency matrix weights on road network heat maps. Our advanced approach enhances the understanding and management of urban traffic systems, making it particularly well-suited for real-world applications.
☆ Adaptive Spatiotemporal Augmentation for Improving Dynamic Graph Learning ICASSP 2025
Dynamic graph augmentation is used to improve the performance of dynamic GNNs. Most methods assume temporal locality, meaning that recent edges are more influential than earlier edges. However, for temporal changes in edges caused by random noise, overemphasizing recent edges while neglecting earlier ones may lead to the model capturing noise. To address this issue, we propose STAA (SpatioTemporal Activity-Aware Random Walk Diffusion). STAA identifies nodes likely to have noisy edges in spatiotemporal dimensions. Spatially, it analyzes critical topological positions through graph wavelet coefficients. Temporally, it analyzes edge evolution through graph wavelet coefficient change rates. Then, random walks are used to reduce the weights of noisy edges, deriving a diffusion matrix containing spatiotemporal information as an augmented adjacency matrix for dynamic GNN learning. Experiments on multiple datasets show that STAA outperforms other dynamic graph augmentation methods in node classification and link prediction tasks.
comment: 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025)
☆ RichSpace: Enriching Text-to-Video Prompt Space via Text Embedding Interpolation
Text-to-video generation models have made impressive progress, but they still struggle with generating videos with complex features. This limitation often arises from the inability of the text encoder to produce accurate embeddings, which hinders the video generation model. In this work, we propose a novel approach to overcome this challenge by selecting the optimal text embedding through interpolation in the embedding space. We demonstrate that this method enables the video generation model to produce the desired videos. Additionally, we introduce a simple algorithm using perpendicular foot embeddings and cosine similarity to identify the optimal interpolation embedding. Our findings highlight the importance of accurate text embeddings and offer a pathway for improving text-to-video generation performance.
☆ Aneumo: A Large-Scale Comprehensive Synthetic Dataset of Aneurysm Hemodynamics
Intracranial aneurysm (IA) is a common cerebrovascular disease that is usually asymptomatic but may cause severe subarachnoid hemorrhage (SAH) if ruptured. Although clinical practice is usually based on individual factors and morphological features of the aneurysm, its pathophysiology and hemodynamic mechanisms remain controversial. To address the limitations of current research, this study constructed a comprehensive hemodynamic dataset of intracranial aneurysms. The dataset is based on 466 real aneurysm models, and 10,000 synthetic models were generated by resection and deformation operations, including 466 aneurysm-free models and 9,534 deformed aneurysm models. The dataset also provides medical image-like segmentation mask files to support insightful analysis. In addition, the dataset contains hemodynamic data measured at eight steady-state flow rates (0.001 to 0.004 kg/s), including critical parameters such as flow velocity, pressure, and wall shear stress, providing a valuable resource for investigating aneurysm pathogenesis and clinical prediction. This dataset will help advance the understanding of the pathologic features and hemodynamic mechanisms of intracranial aneurysms and support in-depth research in related fields. Dataset hosted at https://github.com/Xigui-Li/Aneumo.
☆ Explainable artificial intelligence (XAI): from inherent explainability to large language models
Artificial Intelligence (AI) has continued to achieve tremendous success in recent times. However, the decision logic of these frameworks is often not transparent, making it difficult for stakeholders to understand, interpret or explain their behavior. This limitation hinders trust in machine learning systems and causes a general reluctance towards their adoption in practical applications, particularly in mission-critical domains like healthcare and autonomous driving. Explainable AI (XAI) techniques facilitate the explainability or interpretability of machine learning models, enabling users to discern the basis of the decision and possibly avert undesirable behavior. This comprehensive survey details the advancements of explainable AI methods, from inherently interpretable models to modern approaches for achieving interpretability of various black box models, including large language models (LLMs). Additionally, we review explainable AI techniques that leverage LLM and vision-language model (VLM) frameworks to automate or improve the explainability of other machine learning models. The use of LLM and VLM as interpretability methods particularly enables high-level, semantically meaningful explanations of model decisions and behavior. Throughout the paper, we highlight the scientific principles, strengths and weaknesses of state-of-the-art methods and outline different areas of improvement. Where appropriate, we also present qualitative and quantitative comparison results of various methods to show how they compare. Finally, we discuss the key challenges of XAI and directions for future research.
☆ AIRCHITECT v2: Learning the Hardware Accelerator Design Space through Unified Representations DATE 2025
Design space exploration (DSE) plays a crucial role in enabling custom hardware architectures, particularly for emerging applications like AI, where optimized and specialized designs are essential. With the growing complexity of deep neural networks (DNNs) and the introduction of advanced foundational models (FMs), the design space for DNN accelerators is expanding at an exponential rate. Additionally, this space is highly non-uniform and non-convex, making it increasingly difficult to navigate and optimize. Traditional DSE techniques rely on search-based methods, which involve iterative sampling of the design space to find the optimal solution. However, this process is both time-consuming and often fails to converge to the global optima for such design spaces. Recently, AIrchitect v1, the first attempt to address the limitations of search-based techniques, transformed DSE into a constant-time classification problem using recommendation networks. In this work, we propose AIrchitect v2, a more accurate and generalizable learning-based DSE technique applicable to large-scale design spaces that overcomes the shortcomings of earlier approaches. Specifically, we devise an encoder-decoder transformer model that (a) encodes the complex design space into a uniform intermediate representation using contrastive learning and (b) leverages a novel unified representation blending the advantages of classification and regression to effectively explore the large DSE space without sacrificing accuracy. Experimental results evaluated on 10^5 real DNN workloads demonstrate that, on average, AIrchitect v2 outperforms existing techniques by 15% in identifying optimal design points. Furthermore, to demonstrate the generalizability of our method, we evaluate performance on unseen model workloads (LLMs) and attain a 1.7x improvement in inference latency on the identified hardware architecture.
comment: Accepted to DATE 2025
☆ MultiPruner: Balanced Structure Removal in Foundation Models
Recently, state-of-the-art approaches for pruning large pre-trained models (LPMs) have demonstrated that the training-free removal of non-critical residual blocks in Transformers is viable for reducing model size, achieving results that outperform previous training-free pruning approaches. Motivated by these findings, we extend BlockPruner (Zhong et al., 2024) and propose MultiPruner, a pruning approach that surpasses recent training-free pruning methods by adopting a multidimensional, iterative, fine-grained pruning strategy. In MultiPruner, multidimensional pruning reinstates the structural balance in block-pruned models by sequentially compressing along three dimensions: i) residual blocks, ii) channels of multilayer perceptrons (MLP), and iii) attention heads. This solution enhances zero-shot accuracy on downstream tasks compared to other techniques while improving model compression ratios, producing compressed models with fewer computing and memory requirements. Extensive experiments demonstrate the advantages of the proposed method across various large pre-trained models. The code and pruning configurations are available at https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning.
☆ Client-Centric Federated Adaptive Optimization
Federated Learning (FL) is a distributed learning paradigm where clients collaboratively train a model while keeping their own data private. With an increasing scale of clients and models, FL encounters two key challenges, client drift due to a high degree of statistical/system heterogeneity, and lack of adaptivity. However, most existing FL research is based on unrealistic assumptions that virtually ignore system heterogeneity. In this paper, we propose Client-Centric Federated Adaptive Optimization, which is a class of novel federated adaptive optimization approaches. We enable several features in this framework such as arbitrary client participation, asynchronous server aggregation, and heterogeneous local computing, which are ubiquitous in real-world FL systems but are missed in most existing works. We provide a rigorous convergence analysis of our proposed framework for general nonconvex objectives, which is shown to converge with the best-known rate. Extensive experiments show that our approaches consistently outperform the baseline by a large margin across benchmarks.
☆ A Multi-Scale Feature Extraction and Fusion Deep Learning Method for Classification of Wheat Diseases
Wheat is an important source of dietary fiber and protein that is negatively impacted by a number of risks to its growth. The difficulty of identifying and classifying wheat diseases is discussed with an emphasis on wheat loose smut, leaf rust, and crown and root rot. Addressing conditions like crown and root rot, this study introduces an innovative approach that integrates multi-scale feature extraction with advanced image segmentation techniques to enhance classification accuracy. The proposed method uses neural network models Xception, Inception V3, and ResNet 50 to train on a large wheat disease classification dataset 2020 in conjunction with an ensemble of machine vision classifiers, including voting and stacking. The study shows that the suggested methodology has a superior accuracy of 99.75% in the classification of wheat diseases when compared to current state-of-the-art approaches. A deep learning ensemble model Xception showed the highest accuracy.
☆ HEART: Achieving Timely Multi-Model Training for Vehicle-Edge-Cloud-Integrated Hierarchical Federated Learning
The rapid growth of AI-enabled Internet of Vehicles (IoV) calls for efficient machine learning (ML) solutions that can handle high vehicular mobility and decentralized data. This has motivated the emergence of Hierarchical Federated Learning over vehicle-edge-cloud architectures (VEC-HFL). Nevertheless, one aspect which is underexplored in the literature on VEC-HFL is that vehicles often need to execute multiple ML tasks simultaneously, where this multi-model training environment introduces crucial challenges. First, improper aggregation rules can lead to model obsolescence and prolonged training times. Second, vehicular mobility may result in inefficient data utilization by preventing the vehicles from returning their models to the network edge. Third, achieving a balanced resource allocation across diverse tasks becomes of paramount importance as it majorly affects the effectiveness of collaborative training. We take one of the first steps towards addressing these challenges via proposing a framework for multi-model training in dynamic VEC-HFL with the goal of minimizing global training latency while ensuring balanced training across various tasks-a problem that turns out to be NP-hard. To facilitate timely model training, we introduce a hybrid synchronous-asynchronous aggregation rule. Building on this, we present a novel method called Hybrid Evolutionary And gReedy allocaTion (HEART). The framework operates in two stages: first, it achieves balanced task scheduling through a hybrid heuristic approach that combines improved Particle Swarm Optimization (PSO) and Genetic Algorithms (GA); second, it employs a low-complexity greedy algorithm to determine the training priority of assigned tasks on vehicles. Experiments on real-world datasets demonstrate the superiority of HEART over existing methods.
comment: 14 pages, 6 figures,
☆ Statistical Inference for Sequential Feature Selection after Domain Adaptation
In high-dimensional regression, feature selection methods, such as sequential feature selection (SeqFS), are commonly used to identify relevant features. When data is limited, domain adaptation (DA) becomes crucial for transferring knowledge from a related source domain to a target domain, improving generalization performance. Although SeqFS after DA is an important task in machine learning, none of the existing methods can guarantee the reliability of its results. In this paper, we propose a novel method for testing the features selected by SeqFS-DA. The main advantage of the proposed method is its capability to control the false positive rate (FPR) below a significance level $\alpha$ (e.g., 0.05). Additionally, a strategic approach is introduced to enhance the statistical power of the test. Furthermore, we provide extensions of the proposed method to SeqFS with model selection criteria including AIC, BIC, and adjusted R-squared. Extensive experiments are conducted on both synthetic and real-world datasets to validate the theoretical results and demonstrate the proposed method's superior performance.
☆ Steering Large Language Models with Feature Guided Activation Additions
Effective and reliable control over large language model (LLM) behavior is a significant challenge. While activation steering methods, which add steering vectors to a model's hidden states, are a promising approach, existing techniques often lack precision and interpretability in how they influence model outputs. We introduce Feature Guided Activation Additions (FGAA), a novel activation steering method that leverages insights from Contrastive Activation Addition (CAA) and Sparse Autoencoder-Targeted Steering (SAE-TS). By operating in the latent space of a Sparse Autoencoder (SAE) and employing optimization techniques to select desired SAE features, FGAA constructs precise steering vectors that provide better steering effects while maintaining coherence of steered model outputs. In this regard, evaluations on Gemma-2-2B and Gemma-2-9B models across various steering tasks demonstrate that FGAA outperforms existing steering methods of CAA, SAE decoder steering, and SAE-TS. Our results also highlight important trade-offs between steering scale and general model capabilities that are consistent across all tested steering methods.
comment: 7 maintext pages, 14 appendix pages
☆ Study on a Fast Solver for Combined Field Integral Equations of 3D Conducting Bodies Based on Graph Neural Networks
In this paper, we present a graph neural networks (GNNs)-based fast solver (GraphSolver) for solving combined field integral equations (CFIEs) of 3D conducting bodies. Rao-Wilton-Glisson (RWG) basis functions are employed to discretely and accurately represent the geometry of 3D conducting bodies. A concise and informative graph representation is then constructed by treating each RWG function as a node in the graph, enabling the flow of current between nodes. With the transformed graphs, GraphSolver is developed to directly predict real and imaginary parts of the x, y and z components of the surface current densities at each node (RWG function). Numerical results demonstrate the efficacy of GraphSolver in solving CFIEs for 3D conducting bodies with varying levels of geometric complexity, including basic 3D targets, missile-shaped targets, and airplane-shaped targets.
comment: 10 pages,11 figures
☆ SLIM: Sim-to-Real Legged Instructive Manipulation via Long-Horizon Visuomotor Learning
We present a low-cost quadruped manipulation system that solves long-horizon real-world tasks, trained by reinforcement learning purely in simulation. The system comprises 1) a hierarchical design of a high-level policy for visual-mobile manipulation following instructions, and a low-level policy for quadruped movement and limb-control, 2) a progressive policy expansion approach for solving the long-horizon task together with a teacher-student framework for efficient high-level training of the high-level visuomotor policy, and 3) a suite of techniques for minimizing sim-to-real gaps. With budget-friendly but limited reliability and performance hardware, and just one wrist-mounted RGB camera, the entire system fully trained in simulation achieves high success rates for long horizon tasks involving search, move, grasp, and drop-into, with fluid sim-to-real transfer in a wide variety of indoor and outdoor scenes and lighting conditions.Extensive real-world evaluations show that on the long horizon mobile manipulation tasks, our system achieves good performance when transferred to real both in terms of task success rate and execution efficiency. Finally, we discuss the necessity of our sim-to-real techniques for legged mobile manipulation, and show their ablation performance.
☆ SBAMDT: Bayesian Additive Decision Trees with Adaptive Soft Semi-multivariate Split Rules
Bayesian Additive Regression Trees [BART, Chipman et al., 2010] have gained significant popularity due to their remarkable predictive performance and ability to quantify uncertainty. However, standard decision tree models rely on recursive data splits at each decision node, using deterministic decision rules based on a single univariate feature. This approach limits their ability to effectively capture complex decision boundaries, particularly in scenarios involving multiple features, such as spatial domains, or when transitions are either sharp or smoothly varying. In this paper, we introduce a novel probabilistic additive decision tree model that employs a soft split rule. This method enables highly flexible splits that leverage both univariate and multivariate features, while also respecting the geometric properties of the feature domain. Notably, the probabilistic split rule adapts dynamically across decision nodes, allowing the model to account for varying levels of smoothness in the regression function. We demonstrate the utility of the proposed model through comparisons with existing tree-based models on synthetic datasets and a New York City education dataset.
☆ FoundationStereo: Zero-Shot Stereo Matching
Tremendous progress has been made in deep stereo matching to excel on benchmark datasets through per-domain fine-tuning. However, achieving strong zero-shot generalization - a hallmark of foundation models in other computer vision tasks - remains challenging for stereo matching. We introduce FoundationStereo, a foundation model for stereo depth estimation designed to achieve strong zero-shot generalization. To this end, we first construct a large-scale (1M stereo pairs) synthetic training dataset featuring large diversity and high photorealism, followed by an automatic self-curation pipeline to remove ambiguous samples. We then design a number of network architecture components to enhance scalability, including a side-tuning feature backbone that adapts rich monocular priors from vision foundation models to mitigate the sim-to-real gap, and long-range context reasoning for effective cost volume filtering. Together, these components lead to strong robustness and accuracy across domains, establishing a new standard in zero-shot stereo depth estimation.
☆ Sparse Binary Representation Learning for Knowledge Tracing
Knowledge tracing (KT) models aim to predict students' future performance based on their historical interactions. Most existing KT models rely exclusively on human-defined knowledge concepts (KCs) associated with exercises. As a result, the effectiveness of these models is highly dependent on the quality and completeness of the predefined KCs. Human errors in labeling and the cost of covering all potential underlying KCs can limit model performance. In this paper, we propose a KT model, Sparse Binary Representation KT (SBRKT), that generates new KC labels, referred to as auxiliary KCs, which can augment the predefined KCs to address the limitations of relying solely on human-defined KCs. These are learned through a binary vector representation, where each bit indicates the presence (one) or absence (zero) of an auxiliary KC. The resulting discrete representation allows these auxiliary KCs to be utilized in training any KT model that incorporates KCs. Unlike pre-trained dense embeddings, which are limited to models designed to accept such vectors, our discrete representations are compatible with both classical models, such as Bayesian Knowledge Tracing (BKT), and modern deep learning approaches. To generate this discrete representation, SBRKT employs a binarization method that learns a sparse representation, fully trainable via stochastic gradient descent. Additionally, SBRKT incorporates a recurrent neural network (RNN) to capture temporal dynamics and predict future student responses by effectively combining the auxiliary and predefined KCs. Experimental results demonstrate that SBRKT outperforms the tested baselines on several datasets and achieves competitive performance on others. Furthermore, incorporating the learned auxiliary KCs consistently enhances the performance of BKT across all tested datasets.
♻ ☆ A Fairness-Oriented Reinforcement Learning Approach for the Operation and Control of Shared Micromobility Services
As Machine Learning grows in popularity across various fields, equity has become a key focus for the AI community. However, fairness-oriented approaches are still underexplored in smart mobility. Addressing this gap, our study investigates the balance between performance optimization and algorithmic fairness in shared micromobility services providing a novel framework based on Reinforcement Learning. Exploiting Q-learning, the proposed methodology achieves equitable outcomes in terms of the Gini index across different areas characterized by their distance from central hubs. Through vehicle rebalancing, the provided scheme maximizes operator performance while ensuring fairness principles for users, reducing iniquity by up to 85% while only increasing costs by 30% (w.r.t. applying no equity adjustment). A case study with synthetic data validates our insights and highlights the importance of fairness in urban micromobility (source code: https://github.com/mcederle99/FairMSS.git).
comment: 6 pages, 3 figures, accepted at the 2025 American Control Conference (ACC) on January 17th, 2025
♻ ☆ On Learning Informative Trajectory Embeddings for Imitation, Classification and Regression AAMAS 2025
In real-world sequential decision making tasks like autonomous driving, robotics, and healthcare, learning from observed state-action trajectories is critical for tasks like imitation, classification, and clustering. For example, self-driving cars must replicate human driving behaviors, while robots and healthcare systems benefit from modeling decision sequences, whether or not they come from expert data. Existing trajectory encoding methods often focus on specific tasks or rely on reward signals, limiting their ability to generalize across domains and tasks. Inspired by the success of embedding models like CLIP and BERT in static domains, we propose a novel method for embedding state-action trajectories into a latent space that captures the skills and competencies in the dynamic underlying decision-making processes. This method operates without the need for reward labels, enabling better generalization across diverse domains and tasks. Our contributions are threefold: (1) We introduce a trajectory embedding approach that captures multiple abilities from state-action data. (2) The learned embeddings exhibit strong representational power across downstream tasks, including imitation, classification, clustering, and regression. (3) The embeddings demonstrate unique properties, such as controlling agent behaviors in IQ-Learn and an additive structure in the latent space. Experimental results confirm that our method outperforms traditional approaches, offering more flexible and powerful trajectory representations for various applications. Our code is available at https://github.com/Erasmo1015/vte.
comment: AAMAS 2025
♻ ☆ Stochastic gradient descent for streaming linear and rectified linear systems with adversarial corruptions
We propose SGD-exp, a stochastic gradient descent approach for linear and ReLU regressions under Massart noise (adversarial semi-random corruption model) for the fully streaming setting. We show novel nearly linear convergence guarantees of SGD-exp to the true parameter with up to $50\%$ Massart corruption rate, and with any corruption rate in the case of symmetric oblivious corruptions. This is the first convergence guarantee result for robust ReLU regression in the streaming setting, and it shows the improved convergence rate over previous robust methods for $L_1$ linear regression due to a choice of an exponentially decaying step size, known for its efficiency in practice. Our analysis is based on the drift analysis of a discrete stochastic process, which could also be interesting on its own.
comment: Submitted to a journal
♻ ☆ STPOTR: Simultaneous Human Trajectory and Pose Prediction Using a Non-Autoregressive Transformer for Robot Following Ahead
In this paper, we develop a neural network model to predict future human motion from an observed human motion history. We propose a non-autoregressive transformer architecture to leverage its parallel nature for easier training and fast, accurate predictions at test time. The proposed architecture divides human motion prediction into two parts: 1) the human trajectory, which is the hip joint 3D position over time and 2) the human pose which is the all other joints 3D positions over time with respect to a fixed hip joint. We propose to make the two predictions simultaneously, as the shared representation can improve the model performance. Therefore, the model consists of two sets of encoders and decoders. First, a multi-head attention module applied to encoder outputs improves human trajectory. Second, another multi-head self-attention module applied to encoder outputs concatenated with decoder outputs facilitates learning of temporal dependencies. Our model is well-suited for robotic applications in terms of test accuracy and speed, and compares favorably with respect to state-of-the-art methods. We demonstrate the real-world applicability of our work via the Robot Follow-Ahead task, a challenging yet practical case study for our proposed model.
♻ ☆ The Effect of Similarity Measures on Accurate Stability Estimates for Local Surrogate Models in Text-based Explainable AI
Recent work has investigated the vulnerability of local surrogate methods to adversarial perturbations on a machine learning (ML) model's inputs, where the explanation is manipulated while the meaning and structure of the original input remains similar under the complex model. Although weaknesses across many methods have been shown to exist, the reasons behind why remain little explored. Central to the concept of adversarial attacks on explainable AI (XAI) is the similarity measure used to calculate how one explanation differs from another. A poor choice of similarity measure can lead to erroneous conclusions on the efficacy of an XAI method. Too sensitive a measure results in exaggerated vulnerability, while too coarse understates its weakness. We investigate a variety of similarity measures designed for text-based ranked lists, including Kendall's Tau, Spearman's Footrule, and Rank-biased Overlap to determine how substantial changes in the type of measure or threshold of success affect the conclusions generated from common adversarial attack processes. Certain measures are found to be overly sensitive, resulting in erroneous estimates of stability.
comment: 11 pages, 8 Tables (Minor edits for clarity and grammar)
♻ ☆ High-Rank Irreducible Cartesian Tensor Decomposition and Bases of Equivariant Spaces
Irreducible Cartesian tensors (ICTs) play a crucial role in the design of equivariant graph neural networks, as well as in theoretical chemistry and chemical physics. Meanwhile, the design space of available linear operations on tensors that preserve symmetry presents a significant challenge. The ICT decomposition and a basis of this equivariant space are difficult to obtain for high-rank tensors. After decades of research, Bonvicini (2024) recently achieves an explicit ICT decomposition for $n=5$ with factorial time/space complexity. In this work we, for the first time, obtains decomposition matrices for ICTs up to rank $n=9$ with reduced and affordable complexity, by constructing what we call path matrices. The path matrices are obtained via performing chain-like contractions with Clebsch-Gordan matrices following the parentage scheme. We prove and leverage that the concatenation of path matrices is an orthonormal change-of-basis matrix between the Cartesian tensor product space and the spherical direct sum spaces. Furthermore, we identify a complete orthogonal basis for the equivariant space, rather than a spanning set (Pearce-Crump, 2023), through this path matrices technique. To the best of our knowledge, this is also the first analytic, rather than numerical, method for theoretically obtaining arbitrary rank orthogonal ICT decomposition matrices and orthogonal equivariant bases. We further extend our result to the arbitrary tensor product and direct sum spaces, enabling free design between different spaces while keeping symmetry. The Python code is available at https://github.com/ShihaoShao-GH/ICT-decomposition-and-equivariant-bases, where the $n=6,\dots,9$ ICT decomposition matrices are obtained in 1s, 3s, 11s, and 4m32s on 28-cores Intel(R) Xeon(R) Gold 6330 CPU @ 2.00GHz, respectively.
comment: 47 pages
♻ ☆ Generalized Multi-hop Traffic Pressure for Heterogeneous Traffic Perimeter Control
Perimeter control (PC) prevents loss of traffic network capacity due to congestion in urban areas. Homogeneous PC allows all access points to a protected region to have identical permitted inflow. However, homogeneous PC performs poorly when the congestion in the protected region is heterogeneous (e.g., imbalanced demand) since the homogeneous PC does not consider specific traffic conditions around each perimeter intersection. When the protected region has spatially heterogeneous congestion, one needs to modulate the perimeter inflow rate to be higher near low-density regions and vice versa for high-density regions. A na\"ive approach is to leverage 1-hop traffic pressure to measure traffic condition around perimeter intersections, but such metric is too spatially myopic for PC. To address this issue, we formulate multi-hop downstream pressure grounded on Markov chain theory, which ``looks deeper'' into the protected region beyond perimeter intersections. In addition, we formulate a two-stage hierarchical control scheme that can leverage this novel multi-hop pressure to redistribute the total permitted inflow provided by a pre-trained deep reinforcement learning homogeneous control policy. Experimental results show that our heterogeneous PC approaches leveraging multi-hop pressure significantly outperform homogeneous PC in scenarios where the origin-destination flows are highly imbalanced with high spatial heterogeneity. Moveover, our approach is shown to be robust against turning ratio uncertainties by a sensitivity analysis.
comment: 11 pages main body, 13 figures, journal paper
♻ ☆ Two Types of AI Existential Risk: Decisive and Accumulative
The conventional discourse on existential risks (x-risks) from AI typically focuses on abrupt, dire events caused by advanced AI systems, particularly those that might achieve or surpass human-level intelligence. These events have severe consequences that either lead to human extinction or irreversibly cripple human civilization to a point beyond recovery. This discourse, however, often neglects the serious possibility of AI x-risks manifesting incrementally through a series of smaller yet interconnected disruptions, gradually crossing critical thresholds over time. This paper contrasts the conventional "decisive AI x-risk hypothesis" with an "accumulative AI x-risk hypothesis." While the former envisions an overt AI takeover pathway, characterized by scenarios like uncontrollable superintelligence, the latter suggests a different causal pathway to existential catastrophes. This involves a gradual accumulation of critical AI-induced threats such as severe vulnerabilities and systemic erosion of economic and political structures. The accumulative hypothesis suggests a boiling frog scenario where incremental AI risks slowly converge, undermining societal resilience until a triggering event results in irreversible collapse. Through systems analysis, this paper examines the distinct assumptions differentiating these two hypotheses. It is then argued that the accumulative view can reconcile seemingly incompatible perspectives on AI risks. The implications of differentiating between these causal pathways -- the decisive and the accumulative -- for the governance of AI as well as long-term AI safety are discussed.
comment: Journal article for Philosophical Studies
♻ ☆ Counterfactual Uncertainty Quantification of Factual Estimand of Efficacy from Before-and-After Treatment Repeated Measures Randomized Controlled Trials
The ideal estimand for comparing treatment $Rx$ with a control $C$ is the $\textit{counterfactual}$ efficacy $Rx:C$, the expected differential outcome between $Rx$ and $C$ if each patient were given $\textit{both}$. One hundred years ago, Neyman (1923a) proved unbiased $\textit{point estimation}$ of counterfactual efficacy from designed $\textit{factual}$ experiments is achievable. But he left the determination of how much might the counterfactual variance of this estimate be smaller than the factual variance as an open challenge. This article shows $\textit{counterfactual}$ uncertainty quantification (CUQ), quantifying uncertainty for factual point estimates but in a counterfactual setting, is achievable for Randomized Controlled Trials (RCTs) with Before-and-After treatment Repeated Measures which are common in many therapeutic areas. We achieve CUQ whose variability is typically smaller than factual UQ by creating a new statistical modeling principle called ETZ. We urge caution in using predictors with measurement error which violates standard regression assumption and can cause $\textit{attenuation}$ in estimating treatment effects. Fortunately, we prove that, for traditional medicine in general, and for targeted therapy with efficacy defined as averaged over the population, counterfactual point estimation is unbiased. However, for both Real Human and Digital Twins approaches, predicting treatment effect in $\textit{subgroups}$ may have attenuation bias.
♻ ☆ Automated Machine Learning for Remaining Useful Life Predictions
Being able to predict the remaining useful life (RUL) of an engineering system is an important task in prognostics and health management. Recently, data-driven approaches to RUL predictions are becoming prevalent over model-based approaches since no underlying physical knowledge of the engineering system is required. Yet, this just replaces required expertise of the underlying physics with machine learning (ML) expertise, which is often also not available. Automated machine learning (AutoML) promises to build end-to-end ML pipelines automatically enabling domain experts without ML expertise to create their own models. This paper introduces AutoRUL, an AutoML-driven end-to-end approach for automatic RUL predictions. AutoRUL combines fine-tuned standard regression methods to an ensemble with high predictive power. By evaluating the proposed method on eight real-world and synthetic datasets against state-of-the-art hand-crafted models, we show that AutoML provides a viable alternative to hand-crafted data-driven RUL predictions. Consequently, creating RUL predictions can be made more accessible for domain experts using AutoML by eliminating ML expertise from data-driven model construction.
comment: Manuscript accepted at IEEE SMC 2023
♻ ☆ Enhancing reliability in prediction intervals using point forecasters: Heteroscedastic Quantile Regression and Width-Adaptive Conformal Inference
Constructing prediction intervals for time series forecasting is challenging, particularly when practitioners rely solely on point forecasts. While previous research has focused on creating increasingly efficient intervals, we argue that standard measures alone are inadequate. Beyond efficiency, prediction intervals must adapt their width based on the difficulty of the prediction while preserving coverage regardless of complexity. To address these issues, we propose combining Heteroscedastic Quantile Regression (HQR) with Width-Adaptive Conformal Inference (WACI). This integrated procedure guarantees theoretical coverage and enables interval widths to vary with predictive uncertainty. We assess its performance using both a synthetic example and a real world Electricity Price Forecasting scenario. Our results show that this combined approach meets or surpasses typical benchmarks for validity and efficiency, while also fulfilling important yet often overlooked practical requirements.
♻ ☆ Can machine learning unlock new insights into high-frequency trading?
We design and train machine learning models to capture the nonlinear interactions between financial market dynamics and high-frequency trading (HFT) activity. In doing so, we introduce new metrics to identify liquidity-demanding and -supplying HFT strategies. Both types of HFT strategies increase activity in response to information events and decrease it when trading speed is restricted, with liquidity-supplying strategies demonstrating greater responsiveness. Liquidity-demanding HFT is positively linked with latency arbitrage opportunities, whereas liquidity-supplying HFT is negatively related, aligning with theoretical expectations. Our metrics have implications for understanding the information production process in financial markets.
comment: 66 pages, 6 figures, 11 tables
♻ ☆ Large Language Model is Secretly a Protein Sequence Optimizer
We consider the protein sequence engineering problem, which aims to find protein sequences with high fitness levels, starting from a given wild-type sequence. Directed evolution has been a dominating paradigm in this field which has an iterative process to generate variants and select via experimental feedback. We demonstrate large language models (LLMs), despite being trained on massive texts, are secretly protein sequence optimizers. With a directed evolutionary method, LLM can perform protein engineering through Pareto and experiment-budget constrained optimization, demonstrating success on both synthetic and experimental fitness landscapes.
comment: Preprint
♻ ☆ Bridging Diversity and Uncertainty in Active learning with Self-Supervised Pre-Training ICLR 2024
This study addresses the integration of diversity-based and uncertainty-based sampling strategies in active learning, particularly within the context of self-supervised pre-trained models. We introduce a straightforward heuristic called TCM that mitigates the cold start problem while maintaining strong performance across various data levels. By initially applying TypiClust for diversity sampling and subsequently transitioning to uncertainty sampling with Margin, our approach effectively combines the strengths of both strategies. Our experiments demonstrate that TCM consistently outperforms existing methods across various datasets in both low and high data regimes.
comment: Accepted at ICLR 2024 Workshop on Practical Machine Learning for Low Resource Settings (PML4LRS)
♻ ☆ Optimal Quantization for Matrix Multiplication
Recent work in machine learning community proposed multiple methods for performing lossy compression (quantization) of large matrices. This quantization is important for accelerating matrix multiplication (main component of large language models), which is often bottlenecked by the speed of loading these matrices from memory. Unlike classical vector quantization and rate-distortion theory, the goal of these new compression algorithms is to be able to approximate not the matrices themselves, but their matrix product. Specifically, given a pair of real matrices $A,B$ an encoder (compressor) is applied to each of them independently producing descriptions with $R$ bits per entry. These representations subsequently are used by the decoder to estimate matrix product $A^\top B$. In this work, we provide a non-asymptotic lower bound on the mean squared error of this approximation (as a function of rate $R$) for the case of matrices $A,B$ with iid Gaussian entries. Algorithmically, we construct a universal quantizer based on nested lattices with an explicit guarantee of approximation error for any (non-random) pair of matrices $A$, $B$ in terms of only Frobenius norms $\|\bar{A}\|_F, \|\bar{B}\|_F$ and $\|\bar{A}^\top \bar{B}\|_F$, where $\bar{A},\bar{B}$ are versions of $A,B$ with zero-centered columns, respectively. For iid Gaussian matrices our quantizer achieves the lower bound and is, thus, asymptotically optimal. A practical low-complexity version of our quantizer achieves performance quite close to optimal. In addition, we derive rate-distortion function for matrix multiplication of iid Gaussian matrices, which exhibits an interesting phase-transition at $R\approx 0.906$ bit/entry.
♻ ☆ DPCL-Diff: The Temporal Knowledge Graph Reasoning Based on Graph Node Diffusion Model with Dual-Domain Periodic Contrastive Learning
Temporal knowledge graph (TKG) reasoning that infers future missing facts is an essential and challenging task. Predicting future events typically relies on closely related historical facts, yielding more accurate results for repetitive or periodic events. However, for future events with sparse historical interactions, the effectiveness of this method, which focuses on leveraging high-frequency historical information, diminishes. Recently, the capabilities of diffusion models in image generation have opened new opportunities for TKG reasoning. Therefore, we propose a graph node diffusion model with dual-domain periodic contrastive learning (DPCL-Diff). Graph node diffusion model (GNDiff) introduces noise into sparsely related events to simulate new events, generating high-quality data that better conforms to the actual distribution. This generative mechanism significantly enhances the model's ability to reason about new events. Additionally, the dual-domain periodic contrastive learning (DPCL) maps periodic and non-periodic event entities to Poincar\'e and Euclidean spaces, leveraging their characteristics to distinguish similar periodic events effectively. Experimental results on four public datasets demonstrate that DPCL-Diff significantly outperforms state-of-the-art TKG models in event prediction, demonstrating our approach's effectiveness. This study also investigates the combined effectiveness of GNDiff and DPCL in TKG tasks.
comment: 11 pages, 2 figures
♻ ☆ Jailbreaking as a Reward Misspecification Problem
The widespread adoption of large language models (LLMs) has raised concerns about their safety and reliability, particularly regarding their vulnerability to adversarial attacks. In this paper, we propose a novel perspective that attributes this vulnerability to reward misspecification during the alignment process. This misspecification occurs when the reward function fails to accurately capture the intended behavior, leading to misaligned model outputs. We introduce a metric ReGap to quantify the extent of reward misspecification and demonstrate its effectiveness and robustness in detecting harmful backdoor prompts. Building upon these insights, we present ReMiss, a system for automated red teaming that generates adversarial prompts in a reward-misspecified space. ReMiss achieves state-of-the-art attack success rates on the AdvBench benchmark against various target aligned LLMs while preserving the human readability of the generated prompts. Furthermore, these attacks on open-source models demonstrate high transferability to closed-source models like GPT-4o and out-of-distribution tasks from HarmBench. Detailed analysis highlights the unique advantages of the proposed reward misspecification objective compared to previous methods, offering new insights for improving LLM safety and robustness.
♻ ☆ Boosting drug-disease association prediction for drug repositioning via dual-feature extraction and cross-dual-domain decoding
The extraction of biomedical data has significant academic and practical value in contemporary biomedical sciences. In recent years, drug repositioning, a cost-effective strategy for drug development by discovering new indications for approved drugs, has gained increasing attention. However, many existing drug repositioning methods focus on mining information from adjacent nodes in biomedical networks without considering the potential inter-relationships between the feature spaces of drugs and diseases. This can lead to inaccurate encoding, resulting in biased mined drug-disease association information. To address this limitation, we propose a new model called Dual-Feature Drug Repurposing Neural Network (DFDRNN). DFDRNN allows the mining of two features (similarity and association) from the drug-disease biomedical networks to encode drugs and diseases. A self-attention mechanism is utilized to extract neighbor feature information. It incorporates two dual-feature extraction modules: the single-domain dual-feature extraction (SDDFE) module for extracting features within a single domain (drugs or diseases) and the cross-domain dual-feature extraction (CDDFE) module for extracting features across domains. By utilizing these modules, we ensure more appropriate encoding of drugs and diseases. A cross-dual-domain decoder is also designed to predict drug-disease associations in both domains. Our proposed DFDRNN model outperforms six state-of-the-art methods on four benchmark datasets, achieving an average AUROC of 0.946 and an average AUPR of 0.597. Case studies on two diseases show that the proposed DFDRNN model can be applied in real-world scenarios, demonstrating its significant potential in drug repositioning.
♻ ☆ Bandit on the Hunt: Dynamic Crawling for Cyber Threat Intelligence
Public information contains valuable Cyber Threat Intelligence (CTI) that is used to prevent attacks in the future. Ideally, the learnings from previous attacks help to mitigate all those that follow. While there are standards for sharing this information, much of it is shared in non-standardized news articles or blog posts. It is a time-consuming task to monitor online sources for threats and even then, one can never be sure, to use the right sources. Current research propose extractors of Indicators of Compromise from known sources, while the identification of new sources is rarely considered. This paper proposes a focused crawler focused on the CTI domain based on multi-armed bandit ( MAB) and different crawling strategies. It uses SBERT to identify relevant documents, while dynamically adapt its crawling path. We propose a system called ThreatCrawl, which achieve a harvest rate of over 25% and is able to expand its used seed by over 300%, while retaining focus on the topic at hand. In addition, this crawler identified previously unknown but highly relevant overview pages, datasets, and domains.
comment: 6 pages, 1 figure, 3 tables
♻ ☆ How Redundant Is the Transformer Stack in Speech Representation Models? ICASSP 2025
Self-supervised speech representation models, particularly those leveraging transformer architectures, have demonstrated remarkable performance across various tasks such as speech recognition, speaker identification, and emotion detection. Recent studies on transformer models revealed a high redundancy between layers and the potential for significant pruning, which we will investigate here for transformer-based speech representation models. We perform a detailed analysis of layer similarity in speech representation models using three similarity metrics: cosine similarity, centered kernel alignment, and mutual nearest-neighbor alignment. Our findings reveal a block-like structure of high similarity, suggesting two main processing steps and significant redundancy of layers. We demonstrate the effectiveness of pruning transformer-based speech representation models without the need for post-training, achieving up to 40% reduction in transformer layers while maintaining over 95% of the model's predictive capacity. Furthermore, we employ a knowledge distillation method to substitute the entire transformer stack with mimicking layers, reducing the network size 95-98% and the inference time by up to 94%. This substantial decrease in computational load occurs without considerable performance loss, suggesting that the transformer stack is almost completely redundant for downstream applications of speech representation models.
comment: To appear at ICASSP 2025 (excluding appendix)
♻ ☆ LLM-Based Routing in Mixture of Experts: A Novel Framework for Trading AAAI 2025
Recent advances in deep learning and large language models (LLMs) have facilitated the deployment of the mixture-of-experts (MoE) mechanism in the stock investment domain. While these models have demonstrated promising trading performance, they are often unimodal, neglecting the wealth of information available in other modalities, such as textual data. Moreover, the traditional neural network-based router selection mechanism fails to consider contextual and real-world nuances, resulting in suboptimal expert selection. To address these limitations, we propose LLMoE, a novel framework that employs LLMs as the router within the MoE architecture. Specifically, we replace the conventional neural network-based router with LLMs, leveraging their extensive world knowledge and reasoning capabilities to select experts based on historical price data and stock news. This approach provides a more effective and interpretable selection mechanism. Our experiments on multimodal real-world stock datasets demonstrate that LLMoE outperforms state-of-the-art MoE models and other deep neural network approaches. Additionally, the flexible architecture of LLMoE allows for easy adaptation to various downstream tasks.
comment: Accepted by AAAI 2025 Workshop on AI for Social Impact - Bridging Innovations in Finance, Social Media, and Crime Prevention
♻ ☆ Multi-stage Deep Learning Artifact Reduction for Pallel-beam Computed Tomography
Computed Tomography (CT) using synchrotron radiation is a powerful technique that, compared to lab-CT techniques, boosts high spatial and temporal resolution while also providing access to a range of contrast-formation mechanisms. The acquired projection data is typically processed by a computational pipeline composed of multiple stages. Artifacts introduced during data acquisition can propagate through the pipeline, and degrade image quality in the reconstructed images. Recently, deep learning has shown significant promise in enhancing image quality for images representing scientific data. This success has driven increasing adoption of deep learning techniques in CT imaging. Various approaches have been proposed to incorporate deep learning into computational pipelines, but each has limitations in addressing artifacts effectively and efficiently in synchrotron CT, either in properly addressing the specific artifacts, or in computational efficiency. Recognizing these challenges, we introduce a novel method that incorporates separate deep learning models at each stage of the tomography pipeline-projection, sinogram, and reconstruction-to address specific artifacts locally in a data-driven way. Our approach includes bypass connections that feed both the outputs from previous stages and raw data to subsequent stages, minimizing the risk of error propagation. Extensive evaluations on both simulated and real-world datasets illustrate that our approach effectively reduces artifacts and outperforms comparison methods.
♻ ☆ Annealed Multiple Choice Learning: Overcoming limitations of Winner-takes-all with annealing NeurIPS 2024
We introduce Annealed Multiple Choice Learning (aMCL) which combines simulated annealing with MCL. MCL is a learning framework handling ambiguous tasks by predicting a small set of plausible hypotheses. These hypotheses are trained using the Winner-takes-all (WTA) scheme, which promotes the diversity of the predictions. However, this scheme may converge toward an arbitrarily suboptimal local minimum, due to the greedy nature of WTA. We overcome this limitation using annealing, which enhances the exploration of the hypothesis space during training. We leverage insights from statistical physics and information theory to provide a detailed description of the model training trajectory. Additionally, we validate our algorithm by extensive experiments on synthetic datasets, on the standard UCI benchmark, and on speech separation.
comment: NeurIPS 2024
♻ ☆ LLM360 K2: Building a 65B 360-Open-Source Large Language Model from Scratch
We detail the training of the LLM360 K2-65B model, scaling up our 360-degree OPEN SOURCE approach to the largest and most powerful models under project LLM360. While open-source LLMs continue to advance, the answer to "How are the largest LLMs trained?" remains unclear within the community. The implementation details for such high-capacity models are often protected due to business considerations associated with their high cost. This lack of transparency prevents LLM researchers from leveraging valuable insights from prior experience, e.g., "What are the best practices for addressing loss spikes?" The LLM360 K2 project addresses this gap by providing full transparency and access to resources accumulated during the training of LLMs at the largest scale. This report highlights key elements of the K2 project, including our first model, K2 DIAMOND, a 65 billion-parameter LLM that surpasses LLaMA-65B and rivals LLaMA2-70B, while requiring fewer FLOPs and tokens. We detail the implementation steps and present a longitudinal analysis of K2 DIAMOND's capabilities throughout its training process. We also outline ongoing projects such as TXT360, setting the stage for future models in the series. By offering previously unavailable resources, the K2 project also resonates with the 360-degree OPEN SOURCE principles of transparency, reproducibility, and accessibility, which we believe are vital in the era of resource-intensive AI research.
♻ ☆ BatchLLM: Optimizing Large Batched LLM Inference with Global Prefix Sharing and Throughput-oriented Token Batching
Large language models (LLMs) increasingly play an important role in a wide range of information processing and management tasks. Many of these tasks are performed in large batches or even offline, and the performance indictor for which is throughput. These tasks usually show the characteristic of prefix sharing, where different prompt input can partially show the common prefix. However, the existing LLM inference engines tend to optimize the streaming requests and show limitations of supporting the large batched tasks with the prefix sharing characteristic. The existing solutions use the LRU-based cache to reuse the KV context of common prefix between requests. The KV context that are about to be reused may prematurely evicted with the implicit cache management. Besides, the streaming oriented systems do not leverage the request-batch information and can not mix the decoding tokens with the prefill chunks to the best for the batched scenarios, and thus fails to saturate the GPU. We propose BatchLLM to address the above problems. BatchLLM explicitly identifies the common prefixes globally. The requests sharing the same prefix will be scheduled together to reuse the KV context the best. BatchLLM reorders the requests and schedules the requests with larger ratio of decoding first to better mix the decoding tokens with the latter prefill chunks, and applies memory-centric token batching to enlarge the token-batch sizes, which helps to increase the GPU utilization. Finally, BatchLLM optimizes the prefix-shared Attention kernel with horizontal fusion to reduce tail effect and kernel launch overhead. Extensive evaluation shows that BatchLLM outperforms vLLM and SGLang by 1.3$\times$ to 10.8$\times$ on a set of microbenchmarks and a typical industry workload under different hardware environments.
♻ ☆ ELITR-Bench: A Meeting Assistant Benchmark for Long-Context Language Models COLING 2025
Research on Large Language Models (LLMs) has recently witnessed an increasing interest in extending the models' context size to better capture dependencies within long documents. While benchmarks have been proposed to assess long-range abilities, existing efforts primarily considered generic tasks that are not necessarily aligned with real-world applications. In contrast, we propose a new benchmark for long-context LLMs focused on a practical meeting assistant scenario in which the long contexts consist of transcripts obtained by automatic speech recognition, presenting unique challenges for LLMs due to the inherent noisiness and oral nature of such data. Our benchmark, ELITR-Bench, augments the existing ELITR corpus by adding 271 manually crafted questions with their ground-truth answers, as well as noisy versions of meeting transcripts altered to target different Word Error Rate levels. Our experiments with 12 long-context LLMs on ELITR-Bench confirm the progress made across successive generations of both proprietary and open models, and point out their discrepancies in terms of robustness to transcript noise. We also provide a thorough analysis of our GPT-4-based evaluation, including insights from a crowdsourcing study. Our findings indicate that while GPT-4's scores align with human judges, its ability to distinguish beyond three score levels may be limited.
comment: Published in COLING 2025
♻ ☆ Mitigating analytical variability in fMRI results with style transfer
We propose a novel approach to improve the reproducibility of neuroimaging results by converting statistic maps across different functional MRI pipelines. We make the assumption that pipelines used to compute fMRI statistic maps can be considered as a style component and we propose to use different generative models, among which, Generative Adversarial Networks (GAN) and Diffusion Models (DM) to convert statistic maps across different pipelines. We explore the performance of multiple GAN frameworks, and design a new DM framework for unsupervised multi-domain styletransfer. We constrain the generation of 3D fMRI statistic maps using the latent space of an auxiliary classifier that distinguishes statistic maps from different pipelines and extend traditional sampling techniques used in DM to improve the transition performance. Our experiments demonstrate that our proposed methods aresuccessful: pipelines can indeed be transferred as a style component, providing animportant source of data augmentation for future medical studies.
♻ ☆ Accelerating lensed quasars discovery and modeling with physics-informed variational autoencoders
Strongly lensed quasars provide valuable insights into the rate of cosmic expansion, the distribution of dark matter in foreground deflectors, and the characteristics of quasar hosts. However, detecting them in astronomical images is difficult due to the prevalence of non-lensing objects. To address this challenge, we developed a generative deep learning model called VariLens, built upon a physics-informed variational autoencoder. This model seamlessly integrates three essential modules: image reconstruction, object classification, and lens modeling, offering a fast and comprehensive approach to strong lens analysis. VariLens is capable of rapidly determining both (1) the probability that an object is a lens system and (2) key parameters of a singular isothermal ellipsoid (SIE) mass model -- including the Einstein radius ($\theta_\mathrm{E}$), lens center, and ellipticity -- in just milliseconds using a single CPU. A direct comparison of VariLens estimates with traditional lens modeling for 20 known lensed quasars within the Subaru Hyper Suprime-Cam (HSC) footprint shows good agreement, with both results consistent within $2\sigma$ for systems with $\theta_\mathrm{E}<3$ arcsecs. To identify new lensed quasar candidates, we begin with an initial sample of approximately 80 million sources, combining HSC data with multiwavelength information from various surveys. After applying a photometric preselection aimed at locating $z>1.5$ sources, the number of candidates is reduced to 710,966. Subsequently, VariLens highlights 13,831 sources, each showing a high likelihood of being a lens. A visual assessment of these objects results in 42 promising candidates that await spectroscopic confirmation. These results underscore the potential of automated deep learning pipelines to efficiently detect and model strong lenses in large datasets.
comment: Submitted to the Astronomy & Astrophysics journal and updated to reflect the revised version. The paper consists of 17 main pages, 14 figures, and 5 tables. We welcome feedback and comments from readers!
♻ ☆ VECT-GAN: A variationally encoded generative model for overcoming data scarcity in pharmaceutical science
Data scarcity in pharmaceutical research has led to reliance on labour-intensive trial-and-error approaches for development rather than data-driven methods. While Machine Learning offers a solution, existing datasets are often small and noisy, limiting their utility. To address this, we developed a Variationally Encoded Conditional Tabular Generative Adversarial Network (VECT-GAN), a novel generative model specifically designed for augmenting small, noisy datasets. We introduce a pipeline where data is augmented before regression model development and demonstrate that this consistently and significantly improves performance over other state-of-the-art tabular generative models. We apply this pipeline across six pharmaceutical datasets, and highlight its real-world applicability by developing novel polymers with medically desirable mucoadhesive properties, which we made and experimentally characterised. Additionally, we pre-train the model on the ChEMBL database of drug-like molecules, leveraging knowledge distillation to enhance its generalisability, making it readily available for use on pharmaceutical datasets containing small molecules, an extremely common pharmaceutical task. We demonstrate the power of synthetic data for regularising small tabular datasets, highlighting its potential to become standard practice in pharmaceutical model development, and make our method, including VECT-GAN pre-trained on ChEMBL available as a pip package.
comment: 30 pages, 6 primary figures, 3 supplementary figures
♻ ☆ IterL2Norm: Fast Iterative L2-Normalization
Transformer-based large language models are a memory-bound model whose operation is based on a large amount of data that are marginally reused. Thus, the data movement between a host and accelerator likely dictates the total wall-clock time. Layer normalization is one of the key workloads in the transformer model, following each of multi-head attention and feed-forward network blocks. To reduce data movement, layer normalization needs to be performed on the same chip as the matrix-matrix multiplication engine. To this end, we introduce an iterative L2-normalization method for 1D input (IterL2Norm), ensuring fast convergence to the steady-state solution within five iteration steps and high precision, outperforming the fast inverse square root algorithm in six out of nine cases for FP32 and five out of nine for BFloat16 across the embedding lengths used in the OPT models. Implemented in 32/28nm CMOS, the IterL2Norm macro normalizes $d$-dimensional vectors, where $64 \leq d \leq 1024$, with a latency of 116-227 cycles at 100MHz/1.05V.
comment: Design, Automation & Test in Europe Conference 2025
♻ ☆ Benchmarking Graph Representations and Graph Neural Networks for Multivariate Time Series Classification
Multivariate Time Series Classification (MTSC) enables the analysis if complex temporal data, and thus serves as a cornerstone in various real-world applications, ranging from healthcare to finance. Since the relationship among variables in MTS usually contain crucial cues, a large number of graph-based MTSC approaches have been proposed, as the graph topology and edges can explicitly represent relationships among variables (channels), where not only various MTS graph representation learning strategies but also different Graph Neural Networks (GNNs) have been explored. Despite such progresses, there is no comprehensive study that fairly benchmarks and investigates the performances of existing widely-used graph representation learning strategies/GNN classifiers in the application of different MTSC tasks. In this paper, we present the first benchmark which systematically investigates the effectiveness of the widely-used three node feature definition strategies, four edge feature learning strategies and five GNN architecture, resulting in 60 different variants for graph-based MTSC. These variants are developed and evaluated with a standardized data pipeline and training/validation/testing strategy on 26 widely-used suspensor MTSC datasets. Our experiments highlight that node features significantly influence MTSC performance, while the visualization of edge features illustrates why adaptive edge learning outperforms other edge feature learning methods. The code of the proposed benchmark is publicly available at \url{https://github.com/CVI-yangwn/Benchmark-GNN-for-Multivariate-Time-Series-Classification}.
♻ ☆ Geometric Median (GM) Matching for Robust Data Pruning
Large-scale data collections in the wild, are invariably noisy. Thus developing data pruning strategies that remain robust even in the presence of corruption is critical in practice. In this work, we propose Geometric Median ($\gm$) Matching -- a herding style greedy algorithm that yields a $k$-subset such that the mean of the subset approximates the geometric median of the (potentially) noisy dataset. Theoretically, we show that $\gm$ Matching enjoys an improved $\gO(1/k)$ scaling over $\gO(1/\sqrt{k})$ scaling of uniform sampling; while achieving {\bf optimal breakdown point} of {\bf 1/2} even under {\bf arbitrary} corruption. Extensive experiments across several popular deep learning benchmarks indicate that $\gm$ Matching consistently improves over prior state-of-the-art; the gains become more profound at high rates of corruption and aggressive pruning rates; making $\gm$ Matching a strong baseline for future research in robust data pruning.
♻ ☆ Neural networks for insurance pricing with frequency and severity data: a benchmark study from data preprocessing to technical tariff
Insurers usually turn to generalized linear models for modeling claim frequency and severity data. Due to their success in other fields, machine learning techniques are gaining popularity within the actuarial toolbox. Our paper contributes to the literature on frequency-severity insurance pricing with machine learning via deep learning structures. We present a benchmark study on four insurance data sets with frequency and severity targets in the presence of multiple types of input features. We compare in detail the performance of: a generalized linear model on binned input data, a gradient-boosted tree model, a feed-forward neural network (FFNN), and the combined actuarial neural network (CANN). The CANNs combine a baseline prediction established with a GLM and GBM, respectively, with a neural network correction. We explain the data preprocessing steps with specific focus on the multiple types of input features typically present in tabular insurance data sets, such as postal codes, numeric and categorical covariates. Autoencoders are used to embed the categorical variables into the neural network, and we explore their potential advantages in a frequency-severity setting. Model performance is evaluated not only on out-of-sample deviance but also using statistical and calibration performance criteria and managerial tools to get more nuanced insights. Finally, we construct global surrogate models for the neural nets' frequency and severity models. These surrogates enable the translation of the essential insights captured by the FFNNs or CANNs to GLMs. As such, a technical tariff table results that can easily be deployed in practice.
♻ ☆ Differentially Private Secure Multiplication: Hiding Information in the Rubble of Noise
We consider the problem of private distributed multi-party multiplication. It is well-established that Shamir secret-sharing coding strategies can enable perfect information-theoretic privacy in distributed computation via the celebrated algorithm of Ben Or, Goldwasser and Wigderson (the "BGW algorithm"). However, perfect privacy and accuracy require an honest majority, that is, $N \geq 2t+1$ compute nodes are required to ensure privacy against any $t$ colluding adversarial nodes. By allowing for some controlled amount of information leakage and approximate multiplication instead of exact multiplication, we study coding schemes for the setting where the number of honest nodes can be a minority, that is $N< 2t+1.$ We develop a tight characterization privacy-accuracy trade-off for cases where $N < 2t+1$ by measuring information leakage using {differential} privacy instead of perfect privacy, and using the mean squared error metric for accuracy. A novel technical aspect is an intricately layered noise distribution that merges ideas from differential privacy and Shamir secret-sharing at different layers.
comment: Extended version of papers presented in IEEE ISIT 2022, IEEE ISIT 2023 and TPDP 2023
♻ ☆ RELIEF: Reinforcement Learning Empowered Graph Feature Prompt Tuning KDD 2025
The advent of the "pre-train, prompt" paradigm has recently extended its generalization ability and data efficiency to graph representation learning, following its achievements in Natural Language Processing (NLP). Initial graph prompt tuning approaches tailored specialized prompting functions for Graph Neural Network (GNN) models pre-trained with specific strategies, such as edge prediction, thus limiting their applicability. In contrast, another pioneering line of research has explored universal prompting via adding prompts to the input graph's feature space, thereby removing the reliance on specific pre-training strategies. However, the necessity to add feature prompts to all nodes remains an open question. Motivated by findings from prompt tuning research in the NLP domain, which suggest that highly capable pre-trained models need less conditioning signal to achieve desired behaviors, we advocate for strategically incorporating necessary and lightweight feature prompts to certain graph nodes to enhance downstream task performance. This introduces a combinatorial optimization problem, requiring a policy to decide 1) which nodes to prompt and 2) what specific feature prompts to attach. We then address the problem by framing the prompt incorporation process as a sequential decision-making problem and propose our method, RELIEF, which employs Reinforcement Learning (RL) to optimize it. At each step, the RL agent selects a node (discrete action) and determines the prompt content (continuous action), aiming to maximize cumulative performance gain. Extensive experiments on graph and node-level tasks with various pre-training strategies in few-shot scenarios demonstrate that our RELIEF outperforms fine-tuning and other prompt-based approaches in classification performance and data efficiency. The code is available at https://github.com/JasonZhujp/RELIEF.
comment: Accepted by SIGKDD 2025 (camera-ready version). Due to the space limitation, please refer to the V2 version for more details
♻ ☆ Elucidating the Design Space of Dataset Condensation NeurIPS 2024
Dataset condensation, a concept within data-centric learning, efficiently transfers critical attributes from an original dataset to a synthetic version, maintaining both diversity and realism. This approach significantly improves model training efficiency and is adaptable across multiple application areas. Previous methods in dataset condensation have faced challenges: some incur high computational costs which limit scalability to larger datasets (e.g., MTT, DREAM, and TESLA), while others are restricted to less optimal design spaces, which could hinder potential improvements, especially in smaller datasets (e.g., SRe2L, G-VBSM, and RDED). To address these limitations, we propose a comprehensive design framework that includes specific, effective strategies like implementing soft category-aware matching and adjusting the learning rate schedule. These strategies are grounded in empirical evidence and theoretical backing. Our resulting approach, Elucidate Dataset Condensation (EDC), establishes a benchmark for both small and large-scale dataset condensation. In our testing, EDC achieves state-of-the-art accuracy, reaching 48.6% on ImageNet-1k with a ResNet-18 model at an IPC of 10, which corresponds to a compression ratio of 0.78%. This performance exceeds those of SRe2L, G-VBSM, and RDED by margins of 27.3%, 17.2%, and 6.6%, respectively.
comment: Accepted by NeurIPS 2024
♻ ☆ AceMath: Advancing Frontier Math Reasoning with Post-Training and Reward Modeling
In this paper, we introduce AceMath, a suite of frontier math models that excel in solving complex math problems, along with highly effective reward models capable of evaluating generated solutions and reliably identifying the correct ones. To develop the instruction-tuned math models, we propose a supervised fine-tuning (SFT) process that first achieves competitive performance across general domains, followed by targeted fine-tuning for the math domain using a carefully curated set of prompts and synthetically generated responses. The resulting model, AceMath-72B-Instruct greatly outperforms Qwen2.5-Math-72B-Instruct, GPT-4o and Claude-3.5 Sonnet. To develop math-specialized reward model, we first construct AceMath-RewardBench, a comprehensive and robust benchmark for evaluating math reward models across diverse problems and difficulty levels. After that, we present a systematic approach to build our math reward models. The resulting model, AceMath-72B-RM, consistently outperforms state-of-the-art reward models. Furthermore, when combining AceMath-72B-Instruct with AceMath-72B-RM, we achieve the highest average rm@8 score across the math reasoning benchmarks. We release model weights, training data, and evaluation benchmarks at: https://research.nvidia.com/labs/adlr/acemath
♻ ☆ Harnessing small projectors and multiple views for efficient vision pretraining NeurIPS 2024
Recent progress in self-supervised (SSL) visual representation learning has led to the development of several different proposed frameworks that rely on augmentations of images but use different loss functions. However, there are few theoretically grounded principles to guide practice, so practical implementation of each SSL framework requires several heuristics to achieve competitive performance. In this work, we build on recent analytical results to design practical recommendations for competitive and efficient SSL that are grounded in theory. Specifically, recent theory tells us that existing SSL frameworks are minimizing the same idealized loss, which is to learn features that best match the data similarity kernel defined by the augmentations used. We show how this idealized loss can be reformulated to a functionally equivalent loss that is more efficient to compute. We study the implicit bias of using gradient descent to minimize our reformulated loss function and find that using a stronger orthogonalization constraint with a reduced projector dimensionality should yield good representations. Furthermore, the theory tells us that approximating the reformulated loss should be improved by increasing the number of augmentations, and as such using multiple augmentations should lead to improved convergence. We empirically verify our findings on CIFAR, STL and Imagenet datasets, wherein we demonstrate an improved linear readout performance when training a ResNet-backbone using our theoretically grounded recommendations. Remarkably, we also demonstrate that by leveraging these insights, we can reduce the pretraining dataset size by up to 2$\times$ while maintaining downstream accuracy simply by using more data augmentations. Taken together, our work provides theoretically grounded recommendations that can be used to improve SSL convergence and efficiency.
comment: Accepted to NeurIPS 2024
♻ ☆ Deep Plug-and-Play HIO Approach for Phase Retrieval
In the phase retrieval problem, the aim is the recovery of an unknown image from intensity-only measurements such as Fourier intensity. Although there are several solution approaches, solving this problem is challenging due to its nonlinear and ill-posed nature. Recently, learning-based approaches have emerged as powerful alternatives to the analytical methods for several inverse problems. In the context of phase retrieval, a novel plug-and-play approach that exploits learning-based prior and efficient update steps has been presented at the Computational Optical Sensing and Imaging topical meeting, with demonstrated state-of-the-art performance. The key idea was to incorporate learning-based prior to the Gerchberg-Saxton type algorithms through plug-and-play regularization. In this paper, we present the mathematical development of the method including the derivation of its analytical update steps based on half-quadratic splitting and comparatively evaluate its performance through extensive simulations on a large test dataset. The results show the effectiveness of the method in terms of both image quality, computational efficiency, and robustness to initialization and noise.
comment: 16 pages, 5 figures
♻ ☆ Tree-structured Markov random fields with Poisson marginal distributions
A new family of tree-structured Markov random fields for a vector of discrete counting random variables is introduced. According to the characteristics of the family, the marginal distributions of the Markov random fields are all Poisson with the same mean, and are untied from the strength or structure of their built-in dependence. This key feature is uncommon for Markov random fields and most convenient for applications purposes. The specific properties of this new family confer a straightforward sampling procedure and analytic expressions for the joint probability mass function and the joint probability generating function of the vector of counting random variables, thus granting computational methods that scale well to vectors of high dimension. We study the distribution of the sum of random variables constituting a Markov random field from the proposed family, analyze a random variable's individual contribution to that sum through expected allocations, and establish stochastic orderings to assess a wide understanding of their behavior.
comment: 27 pages, 10 figures
♻ ☆ Instruction-Guided Fusion of Multi-Layer Visual Features in Large Vision-Language Models
Large Vision-Language Models (LVLMs) have achieved remarkable success in a wide range of multimodal tasks by integrating pre-trained vision encoders and large language models. However, current LVLMs primarily rely on visual features extracted from the final layers of the vision encoder, overlooking the complementary information available in shallower layers. While recent approaches have explored the use of multilayer visual features in LVLMs, they tend to be task-agnostic and fail to examine the dependencies of hierarchical visual features on specific tasks. To address these gaps, we systematically investigate the contributions of visual features from different encoder layers using 18 benchmarks spanning 6 task categories. Our findings reveal that multilayer features provide complementary strengths with varying task dependencies, and uniform fusion leads to suboptimal performance. Building on these insights, we propose the instruction-guided vision aggregator, a module that dynamically integrates multi-layer visual features based on textual instructions, without increasing the number of visual tokens. Extensive evaluations demonstrate the superior performance of our method. Additionally, an in-depth analysis of the aggregator's behavior highlights the dominance of mid-to-high-level features in semantic-rich tasks and the critical role of low-level features in fine-grained perception.
♻ ☆ TraceFL: Interpretability-Driven Debugging in Federated Learning via Neuron Provenance ICSE
In Federated Learning, clients train models on local data and send updates to a central server, which aggregates them into a global model using a fusion algorithm. This collaborative yet privacy-preserving training comes at a cost. FL developers face significant challenges in attributing global model predictions to specific clients. Localizing responsible clients is a crucial step towards (a) excluding clients primarily responsible for incorrect predictions and (b) encouraging clients who contributed high-quality models to continue participating in the future. Existing ML debugging approaches are inherently inapplicable as they are designed for single-model, centralized training. We introduce TraceFL, a fine-grained neuron provenance capturing mechanism that identifies clients responsible for a global model's prediction by tracking the flow of information from individual clients to the global model. Since inference on different inputs activates a different set of neurons of the global model, TraceFL dynamically quantifies the significance of the global model's neurons in a given prediction, identifying the most crucial neurons in the global model. It then maps them to the corresponding neurons in every participating client to determine each client's contribution, ultimately localizing the responsible client. We evaluate TraceFL on six datasets, including two real-world medical imaging datasets and four neural networks, including advanced models such as GPT. TraceFL achieves 99% accuracy in localizing the responsible client in FL tasks spanning both image and text classification tasks. At a time when state-of-the-artML debugging approaches are mostly domain-specific (e.g., image classification only), TraceFL is the first technique to enable highly accurate automated reasoning across a wide range of FL applications.
comment: Accepted at 2025 IEEE/ACM 47th International Conference on Software Engineering (ICSE)
♻ ☆ Learning Dynamical Systems by Leveraging Data from Similar Systems
We consider the problem of learning the dynamics of a linear system when one has access to data generated by an auxiliary system that shares similar (but not identical) dynamics, in addition to data from the true system. We use a weighted least squares approach, and provide finite sample error bounds of the learned model as a function of the number of samples and various system parameters from the two systems as well as the weight assigned to the auxiliary data. We show that the auxiliary data can help to reduce the intrinsic system identification error due to noise, at the price of adding a portion of error that is due to the differences between the two system models. We further provide a data-dependent bound that is computable when some prior knowledge about the systems, such as upper bounds on noise levels and model difference, is available. This bound can also be used to determine the weight that should be assigned to the auxiliary data during the model training stage.
comment: 15 pages,9 figures
♻ ☆ The Spatial Complexity of Optical Computing and How to Reduce It
Similar to algorithms, which consume time and memory to run, hardware requires resources to function. For devices processing physical waves, implementing operations needs sufficient "space," as dictated by wave physics. How much space is needed to perform a certain function is a fundamental question in optics, with recent research addressing it for given mathematical operations, but not for more general computing tasks, e.g., classification. Inspired by computational complexity theory, we study the "spatial complexity" of optical computing systems in terms of scaling laws - specifically, how their physical dimensions must scale as the dimension of the mathematical operation increases - and propose a new paradigm for designing optical computing systems: space-efficient neuromorphic optics, based on structural sparsity constraints and neural pruning methods motivated by wave physics (notably, the concept of "overlapping nonlocality"). On two mainstream platforms, free-space optics and on-chip integrated photonics, our methods demonstrate substantial size reductions (to 1%-10% the size of conventional designs) with minimal compromise on performance. Our theoretical and computational results reveal a trend of diminishing returns on accuracy as structure dimensions increase, providing a new perspective for interpreting and approaching the ultimate limits of optical computing - a balanced trade-off between device size and accuracy.
♻ ☆ Consistent estimation of generative model representations in the data kernel perspective space
Generative models, such as large language models and text-to-image diffusion models, produce relevant information when presented a query. Different models may produce different information when presented the same query. As the landscape of generative models evolves, it is important to develop techniques to study and analyze differences in model behaviour. In this paper we present novel theoretical results for embedding-based representations of generative models in the context of a set of queries. In particular, we establish sufficient conditions for the consistent estimation of the model embeddings in situations where the query set and the number of models grow.
♻ ☆ Can AI-Generated Text be Reliably Detected?
Large Language Models (LLMs) perform impressively well in various applications. However, the potential for misuse of these models in activities such as plagiarism, generating fake news, and spamming has raised concern about their responsible use. Consequently, the reliable detection of AI-generated text has become a critical area of research. AI text detectors have shown to be effective under their specific settings. In this paper, we stress-test the robustness of these AI text detectors in the presence of an attacker. We introduce recursive paraphrasing attack to stress test a wide range of detection schemes, including the ones using the watermarking as well as neural network-based detectors, zero shot classifiers, and retrieval-based detectors. Our experiments conducted on passages, each approximately 300 tokens long, reveal the varying sensitivities of these detectors to our attacks. Our findings indicate that while our recursive paraphrasing method can significantly reduce detection rates, it only slightly degrades text quality in many cases, highlighting potential vulnerabilities in current detection systems in the presence of an attacker. Additionally, we investigate the susceptibility of watermarked LLMs to spoofing attacks aimed at misclassifying human-written text as AI-generated. We demonstrate that an attacker can infer hidden AI text signatures without white-box access to the detection method, potentially leading to reputational risks for LLM developers. Finally, we provide a theoretical framework connecting the AUROC of the best possible detector to the Total Variation distance between human and AI text distributions. This analysis offers insights into the fundamental challenges of reliable detection as language models continue to advance. Our code is publicly available at https://github.com/vinusankars/Reliability-of-AI-text-detectors.
comment: Published in Transactions on Machine Learning Research (TMLR)
♻ ☆ Enhancing User Interest based on Stream Clustering and Memory Networks in Large-Scale Recommender Systems
Recommender Systems (RSs) provide personalized recommendation service based on user interest, which are widely used in various platforms. However, there are lots of users with sparse interest due to lacking consumption behaviors, which leads to poor recommendation results for them. This problem is widespread in large-scale RSs and is particularly difficult to address. To solve this problem, we propose a novel solution named User Interest Enhancement (UIE) which enhances user interest including user profile and user history behavior sequences using the enhancement vectors and personalized enhancement vector generated based on stream clustering and memory networks from different perspectives. UIE not only remarkably improves model performance on the users with sparse interest but also significantly enhance model performance on other users. UIE is an end-to-end solution which is easy to be implemented based on ranking model. Moreover, we expand our solution and apply similar methods to long-tail items, which also achieves excellent improvement. Furthermore, we conduct extensive offline and online experiments in a large-scale industrial RS. The results demonstrate that our model outperforms other models remarkably, especially for the users with sparse interest. Until now, UIE has been fully deployed in multiple large-scale RSs and achieved remarkable improvements.
♻ ☆ Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluators
Large Language Models (LLMs) have demonstrated promising capabilities as automatic evaluators in assessing the quality of generated natural language. However, LLMs still exhibit biases in evaluation and often struggle to generate coherent evaluations that align with human assessments. In this work, we first conduct a systematic study of the misalignment between LLM evaluators and human evaluation, revealing that existing calibration methods aimed at mitigating biases of LLMs are insufficient for effectively aligning LLM evaluators. Inspired by the use of preference data in RLHF, we formulate the evaluation as a ranking problem and introduce Pairwise-preference Search (PAIRS), an uncertainty-guided search-based rank aggregation method that employs LLMs to conduct pairwise comparisons locally and efficiently ranks candidate texts globally. PAIRS achieves state-of-the-art performance on representative evaluation tasks in long-form generations and demonstrates significant improvements over direct scoring. Furthermore, we provide insights into the role of pairwise preference in quantifying the transitivity of LLMs and demonstrate how PAIRS benefits from calibration using debiased pairwise evaluations.
comment: This paper has been accepted by COLM 2024
♻ ☆ Fast Matrix Multiplications for Lookup Table-Quantized LLMs EMNLP 2024
The deployment of large language models (LLMs) is often constrained by memory bandwidth, where the primary bottleneck is the cost of transferring model parameters from the GPU's global memory to its registers. When coupled with custom kernels that fuse the dequantization and matmul operations, weight-only quantization can thus enable faster inference by reducing the amount of memory movement. However, developing high-performance kernels for weight-quantized LLMs presents substantial challenges, especially when the weights are compressed to non-evenly-divisible bit widths (e.g., 3 bits) with non-uniform, lookup table (LUT) quantization. This paper describes FLUTE, a flexible lookup table engine for LUT-quantized LLMs, which uses offline restructuring of the quantized weight matrix to minimize bit manipulations associated with unpacking, and vectorization and duplication of the lookup table to mitigate shared memory bandwidth constraints. At batch sizes < 32 and quantization group size of 128 (typical in LLM inference), the FLUTE kernel can be 2-4x faster than existing GEMM kernels. As an application of FLUTE, we explore a simple extension to lookup table-based NormalFloat quantization and apply it to quantize LLaMA3 to various configurations, obtaining competitive quantization performance against strong baselines while obtaining an end-to-end throughput increase of 1.5 to 2 times.
comment: EMNLP 2024 (Findings)
♻ ☆ Spatial Clustering of Citizen Science Data Improves Downstream Species Distribution Models AAAI 2025
Citizen science biodiversity data present great opportunities for ecology and conservation across vast spatial and temporal scales. However, the opportunistic nature of these data lacks the sampling structure required by modeling methodologies that address a pervasive challenge in ecological data collection: imperfect detection, i.e., the likelihood of under-observing species on field surveys. Occupancy modeling is an example of an approach that accounts for imperfect detection by explicitly modeling the observation process separately from the biological process of habitat selection. This produces species distribution models that speak to the pattern of the species on a landscape after accounting for imperfect detection in the data, rather than the pattern of species observations corrupted by errors. To achieve this benefit, occupancy models require multiple surveys of a site across which the site's status (i.e., occupied or not) is assumed constant. Since citizen science data are not collected under the required repeated-visit protocol, observations may be grouped into sites post hoc. Existing approaches for constructing sites discard some observations and/or consider only geographic distance and not environmental similarity. In this study, we compare ten approaches for site construction in terms of their impact on downstream species distribution models for 31 bird species in Oregon, using observations recorded in the eBird database. We find that occupancy models built on sites constructed by spatial clustering algorithms perform better than existing alternatives.
comment: AAAI 2025
♻ ☆ Decoupled Sequence and Structure Generation for Realistic Antibody Design
Recently, deep learning has made rapid progress in antibody design, which plays a key role in the advancement of therapeutics. A dominant paradigm is to train a model to jointly generate the antibody sequence and the structure as a candidate. However, the joint generation requires the model to generate both the discrete amino acid categories and the continuous 3D coordinates; this limits the space of possible architectures and may lead to suboptimal performance. In response, we propose an antibody sequence-structure decoupling (ASSD) framework, which separates sequence generation and structure prediction. Although our approach is simple, our idea allows the use of powerful neural architectures and demonstrates notable performance improvements. We also find that the widely used non-autoregressive generators promote sequences with overly repeating tokens. Such sequences are both out-of-distribution and prone to undesirable developability properties that can trigger harmful immune responses in patients. To resolve this, we introduce a composition-based objective that allows an efficient trade-off between high performance and low token repetition. ASSD shows improved performance in various antibody design experiments, while the composition-based objective successfully mitigates token repetition of non-autoregressive models.
comment: 22 pages, 6 figures
♻ ☆ Bayesian Adaptive Calibration and Optimal Design NeurIPS 2024
The process of calibrating computer models of natural phenomena is essential for applications in the physical sciences, where plenty of domain knowledge can be embedded into simulations and then calibrated against real observations. Current machine learning approaches, however, mostly rely on rerunning simulations over a fixed set of designs available in the observed data, potentially neglecting informative correlations across the design space and requiring a large amount of simulations. Instead, we consider the calibration process from the perspective of Bayesian adaptive experimental design and propose a data-efficient algorithm to run maximally informative simulations within a batch-sequential process. At each round, the algorithm jointly estimates the parameters of the posterior distribution and optimal designs by maximising a variational lower bound of the expected information gain. The simulator is modelled as a sample from a Gaussian process, which allows us to correlate simulations and observed data with the unknown calibration parameters. We show the benefits of our method when compared to related approaches across synthetic and real-data problems.
comment: NeurIPS 2024 final revision
♻ ☆ Keeping LLMs Aligned After Fine-tuning: The Crucial Role of Prompt Templates NeurIPS 2024
Public LLMs such as the Llama 2-Chat underwent alignment training and were considered safe. Recently Qi et al. [2024] reported that even benign fine-tuning on seemingly safe datasets can give rise to unsafe behaviors in the models. The current paper is about methods and best practices to mitigate such loss of alignment. We focus on the setting where a public model is fine-tuned before serving users for specific usage, where the model should improve on the downstream task while maintaining alignment. Through extensive experiments on several chat models (Meta's Llama 2-Chat, Mistral AI's Mistral 7B Instruct v0.2, and OpenAI's GPT-3.5 Turbo), this paper uncovers that the prompt templates used during fine-tuning and inference play a crucial role in preserving safety alignment, and proposes the ``Pure Tuning, Safe Testing'' (PTST) strategy -- fine-tune models without a safety prompt, but include it at test time. This seemingly counterintuitive strategy incorporates an intended distribution shift to encourage alignment preservation. Fine-tuning experiments on GSM8K, ChatDoctor, and OpenOrca show that PTST significantly reduces the rise of unsafe behaviors.
comment: NeurIPS 2024
♻ ☆ RLPF: Reinforcement Learning from Prediction Feedback for User Summarization with LLMs AAAI 2025
LLM-powered personalization agent systems employ Large Language Models (LLMs) to predict users' behavior from their past activities. However, their effectiveness often hinges on the ability to effectively leverage extensive, long user historical data due to its inherent noise and length of such data. Existing pretrained LLMs may generate summaries that are concise but lack the necessary context for downstream tasks, hindering their utility in personalization systems. To address these challenges, we introduce Reinforcement Learning from Prediction Feedback (RLPF). RLPF fine-tunes LLMs to generate concise, human-readable user summaries that are optimized for downstream task performance. By maximizing the usefulness of the generated summaries, RLPF effectively distills extensive user history data while preserving essential information for downstream tasks. Our empirical evaluation demonstrates significant improvements in both extrinsic downstream task utility and intrinsic summary quality, surpassing baseline methods by up to 22% on downstream task performance and achieving an up to 84.59% win rate on Factuality, Abstractiveness, and Readability. RLPF also achieves a remarkable 74% reduction in context length while improving performance on 16 out of 19 unseen tasks and/or datasets, showcasing its generalizability. This approach offers a promising solution for enhancing LLM personalization by effectively transforming long, noisy user histories into informative and human-readable representations.
comment: AAAI 2025
♻ ☆ A Systematic Study of Multi-Agent Deep Reinforcement Learning for Safe and Robust Autonomous Highway Ramp Entry
Vehicles today can drive themselves on highways and driverless robotaxis operate in major cities, with more sophisticated levels of autonomous driving expected to be available and become more common in the future. Yet, technically speaking, so-called "Level 5" (L5) operation, corresponding to full autonomy, has not been achieved. For that to happen, functions such as fully autonomous highway ramp entry must be available, and provide provably safe, and reliably robust behavior to enable full autonomy. We present a systematic study of a highway ramp function that controls the vehicles forward-moving actions to minimize collisions with the stream of highway traffic into which a merging (ego) vehicle enters. We take a game-theoretic multi-agent (MA) approach to this problem and study the use of controllers based on deep reinforcement learning (DRL). The virtual environment of the MA DRL uses self-play with simulated data where merging vehicles safely learn to control longitudinal position during a taper-type merge. The work presented in this paper extends existing work by studying the interaction of more than two vehicles (agents) and does so by systematically expanding the road scene with additional traffic and ego vehicles. While previous work on the two-vehicle setting established that collision-free controllers are theoretically impossible in fully decentralized, non-coordinated environments, we empirically show that controllers learned using our approach are nearly ideal when measured against idealized optimal controllers.
comment: 9 pages, 9 figures; added support ack
♻ ☆ A Complete Characterization of Learnability for Stochastic Noisy Bandits
We study the stochastic noisy bandit problem with an unknown reward function $f^*$ in a known function class $\mathcal{F}$. Formally, a model $M$ maps arms $\pi$ to a probability distribution $M(\pi)$ of reward. A model class $\mathcal{M}$ is a collection of models. For each model $M$, define its mean reward function $f^M(\pi)=\mathbb{E}_{r \sim M(\pi)}[r]$. In the bandit learning problem, we proceed in rounds, pulling one arm $\pi$ each round and observing a reward sampled from $M(\pi)$. With knowledge of $\mathcal{M}$, supposing that the true model $M\in \mathcal{M}$, the objective is to identify an arm $\hat{\pi}$ of near-maximal mean reward $f^M(\hat{\pi})$ with high probability in a bounded number of rounds. If this is possible, then the model class is said to be learnable. Importantly, a result of \cite{hanneke2023bandit} shows there exist model classes for which learnability is undecidable. However, the model class they consider features deterministic rewards, and they raise the question of whether learnability is decidable for classes containing sufficiently noisy models. For the first time, we answer this question in the positive by giving a complete characterization of learnability for model classes with arbitrary noise. In addition to that, we also describe the full spectrum of possible optimal query complexities. Further, we prove adaptivity is sometimes necessary to achieve the optimal query complexity. Last, we revisit an important complexity measure for interactive decision making, the Decision-Estimation-Coefficient \citep{foster2021statistical,foster2023tight}, and propose a new variant of the DEC which also characterizes learnability in this setting.
Artificial Intelligence 1
☆ 3rd Workshop on Maritime Computer Vision (MaCVi) 2025: Challenge Results
The 3rd Workshop on Maritime Computer Vision (MaCVi) 2025 addresses maritime computer vision for Unmanned Surface Vehicles (USV) and underwater. This report offers a comprehensive overview of the findings from the challenges. We provide both statistical and qualitative analyses, evaluating trends from over 700 submissions. All datasets, evaluation code, and the leaderboard are available to the public at https://macvi.org/workshop/macvi25.
comment: Part of the MaCVi 2025 workshop
Robotics 20
☆ DexForce: Extracting Force-informed Actions from Kinesthetic Demonstrations for Dexterous Manipulation
Imitation learning requires high-quality demonstrations consisting of sequences of state-action pairs. For contact-rich dexterous manipulation tasks that require fine-grained dexterity, the actions in these state-action pairs must produce the right forces. Current widely-used methods for collecting dexterous manipulation demonstrations are difficult to use for demonstrating contact-rich tasks due to unintuitive human-to-robot motion retargeting and the lack of direct haptic feedback. Motivated by this, we propose DexForce, a method for collecting demonstrations of contact-rich dexterous manipulation. DexForce leverages contact forces, measured during kinesthetic demonstrations, to compute force-informed actions for policy learning. We use DexForce to collect demonstrations for six tasks and show that policies trained on our force-informed actions achieve an average success rate of 76% across all tasks. In contrast, policies trained directly on actions that do not account for contact forces have near-zero success rates. We also conduct a study ablating the inclusion of force data in policy observations. We find that while using force data never hurts policy performance, it helps the most for tasks that require an advanced level of precision and coordination, like opening an AirPods case and unscrewing a nut.
comment: Videos can be found here: https://clairelc.github.io/dexforce.github.io/
☆ Deployment of an Aerial Multi-agent System for Automated Task Execution in Large-scale Underground Mining Environments
In this article, we present a framework for deploying an aerial multi-agent system in large-scale subterranean environments with minimal infrastructure for supporting multi-agent operations. The multi-agent objective is to optimally and reactively allocate and execute inspection tasks in a mine, which are entered by a mine operator on-the-fly. The assignment of currently available tasks to the team of agents is accomplished through an auction-based system, where the agents bid for available tasks, which are used by a central auctioneer to optimally assigns tasks to agents. A mobile Wi-Fi mesh supports inter-agent communication and bi-directional communication between the agents and the task allocator, while the task execution is performed completely infrastructure-free. Given a task to be accomplished, a reliable and modular agent behavior is synthesized by generating behavior trees from a pool of agent capabilities, using a back-chaining approach. The auction system in the proposed framework is reactive and supports addition of new operator-specified tasks on-the-go, at any point through a user-friendly operator interface. The framework has been validated in a real underground mining environment using three aerial agents, with several inspection locations spread in an environment of almost 200 meters. The proposed framework can be utilized for missions involving rapid inspection, gas detection, distributed sensing and mapping etc. in a subterranean environment. The proposed framework and its field deployment contributes towards furthering reliable automation in large-scale subterranean environments to offload both routine and dangerous tasks from human operators to autonomous aerial robots.
comment: Submitted to IEEE Transactions on Field Robotics
☆ Tethered Variable Inertial Attitude Control Mechanisms through a Modular Jumping Limbed Robot
This paper presents the concept of a tethered variable inertial attitude control mechanism for a modular jumping-limbed robot designed for planetary exploration in low-gravity environments. The system, named SPLITTER, comprises two sub-10 kg quadrupedal robots connected by a tether, capable of executing successive jumping gaits and stabilizing in-flight using inertial morphing technology. Through model predictive control (MPC), attitude control was demonstrated by adjusting the limbs and tether length to modulate the system's principal moments of inertia. Our results indicate that this control strategy allows the robot to stabilize during flight phases without needing traditional flywheel-based systems or relying on aerodynamics, making the approach mass-efficient and ideal for small-scale planetary robots' successive jumps. The paper outlines the dynamics, MPC formulation for inertial morphing, actuator requirements, and simulation results, illustrating the potential of agile exploration for small-scale rovers in low-gravity environments like the Moon or asteroids.
comment: Proceeding to IEEE Aerospace Conference 2025
☆ Universal Actions for Enhanced Embodied Foundation Models
Training on diverse, internet-scale data is a key factor in the success of recent large foundation models. Yet, using the same recipe for building embodied agents has faced noticeable difficulties. Despite the availability of many crowd-sourced embodied datasets, their action spaces often exhibit significant heterogeneity due to distinct physical embodiment and control interfaces for different robots, causing substantial challenges in developing embodied foundation models using cross-domain data. In this paper, we introduce UniAct, a new embodied foundation modeling framework operating in a tokenized Universal Action Space. Our learned universal actions capture the generic atomic behaviors across diverse robots by exploiting their shared structural features, and enable enhanced cross-domain data utilization and cross-embodiment generalizations by eliminating the notorious heterogeneity. The universal actions can be efficiently translated back to heterogeneous actionable commands by simply adding embodiment-specific details, from which fast adaptation to new robots becomes simple and straightforward. Our 0.5B instantiation of UniAct outperforms 14X larger SOTA embodied foundation models in extensive evaluations on various real-world and simulation robots, showcasing exceptional cross-embodiment control and adaptation capability, highlighting the crucial benefit of adopting universal actions. Project page: https://github.com/2toinf/UniAct
comment: Preprint
☆ Robotic World Model: A Neural Network Simulator for Robust Policy Optimization in Robotics
Learning robust and generalizable world models is crucial for enabling efficient and scalable robotic control in real-world environments. In this work, we introduce a novel framework for learning world models that accurately capture complex, partially observable, and stochastic dynamics. The proposed method employs a dual-autoregressive mechanism and self-supervised training to achieve reliable long-horizon predictions without relying on domain-specific inductive biases, ensuring adaptability across diverse robotic tasks. We further propose a policy optimization framework that leverages world models for efficient training in imagined environments and seamless deployment in real-world systems. Through extensive experiments, our approach consistently outperforms state-of-the-art methods, demonstrating superior autoregressive prediction accuracy, robustness to noise, and generalization across manipulation and locomotion tasks. Notably, policies trained with our method are successfully deployed on ANYmal D hardware in a zero-shot transfer, achieving robust performance with minimal sim-to-real performance loss. This work advances model-based reinforcement learning by addressing the challenges of long-horizon prediction, error accumulation, and sim-to-real transfer. By providing a scalable and robust framework, the introduced methods pave the way for adaptive and efficient robotic systems in real-world applications.
☆ SpatialCoT: Advancing Spatial Reasoning through Coordinate Alignment and Chain-of-Thought for Embodied Task Planning
Spatial reasoning is an essential problem in embodied AI research. Efforts to enhance spatial reasoning abilities through supplementary spatial data and fine-tuning have proven limited and ineffective when addressing complex embodied tasks, largely due to their dependence on language-based outputs. While some approaches have introduced a point-based action space to mitigate this issue, they fall short in managing more intricate tasks within complex environments. This deficiency arises from their failure to fully exploit the inherent thinking and reasoning capabilities that are fundamental strengths of Vision-Language Models (VLMs). To address these limitations, we propose a novel approach named SpatialCoT, specifically designed to bolster the spatial reasoning capabilities of VLMs. Our approach comprises two stages: spatial coordinate bi-directional alignment, which aligns vision-language inputs with spatial coordinates, and chain-of-thought spatial grounding, which harnesses the reasoning capabilities of language models for advanced spatial reasoning. We evaluate SpatialCoT on challenging navigation and manipulation tasks, both in simulation and real-world settings. Experimental results demonstrate that our method significantly outperforms previous state-of-the-art approaches in both tasks.
comment: 13 pages, 6 figures
☆ A Comprehensive Insights into Drones: History, Classification, Architecture, Navigation, Applications, Challenges, and Future Trends
Unmanned Aerial Vehicles (UAVs), commonly known as Drones, are one of 21st century most transformative technologies. Emerging first for military use, advancements in materials, electronics, and software have catapulted drones into multipurpose tools for a wide range of industries. In this paper, we have covered the history, taxonomy, architecture, navigation systems and branched activities for the same. It explores important future trends like autonomous navigation, AI integration, and obstacle avoidance systems, emphasizing how they contribute to improving the efficiency and versatility of drones. It also looks at the major challenges like technical, environmental, economic, regulatory and ethical, that limit the actual take-up of drones, as well as trends that are likely to mitigate these obstacles in the future. This work offers a structured synthesis of existing studies and perspectives that enable insights about how drones will transform agriculture, logistics, healthcare, disaster management, and other areas, while also identifying new opportunities for innovation and development.
☆ Adaptive Twisting Sliding Control for Integrated Attack UAV's Autopilot and Guidance
This paper investigates an adaptive sliding-mode control for an integrated UAV autopilot and guidance system. First, a two-dimensional mathematical model of the system is derived by considering the incorporated lateral dynamics and relative kinematics of the UAV and its potential target of attack. Then, a sliding surface is derived utilizing the zero-effort miss distance. An adaptive twisting sliding mode (ATSMC) algorithm is applied to the integrated system. Simulation and comparisons have been accomplished. The results show our proposed design performs well in interception precision, even with high nonlinearity, uncertainties, disturbances, and abrupt changes in the target's movement, thanks to the adaptation strategy.
comment: in Proceedings of the 2025 International Conference on Energy, Infrastructure and Environmental Research (EIER2025)
☆ SLIM: Sim-to-Real Legged Instructive Manipulation via Long-Horizon Visuomotor Learning
We present a low-cost quadruped manipulation system that solves long-horizon real-world tasks, trained by reinforcement learning purely in simulation. The system comprises 1) a hierarchical design of a high-level policy for visual-mobile manipulation following instructions, and a low-level policy for quadruped movement and limb-control, 2) a progressive policy expansion approach for solving the long-horizon task together with a teacher-student framework for efficient high-level training of the high-level visuomotor policy, and 3) a suite of techniques for minimizing sim-to-real gaps. With budget-friendly but limited reliability and performance hardware, and just one wrist-mounted RGB camera, the entire system fully trained in simulation achieves high success rates for long horizon tasks involving search, move, grasp, and drop-into, with fluid sim-to-real transfer in a wide variety of indoor and outdoor scenes and lighting conditions.Extensive real-world evaluations show that on the long horizon mobile manipulation tasks, our system achieves good performance when transferred to real both in terms of task success rate and execution efficiency. Finally, we discuss the necessity of our sim-to-real techniques for legged mobile manipulation, and show their ablation performance.
☆ FoundationStereo: Zero-Shot Stereo Matching
Tremendous progress has been made in deep stereo matching to excel on benchmark datasets through per-domain fine-tuning. However, achieving strong zero-shot generalization - a hallmark of foundation models in other computer vision tasks - remains challenging for stereo matching. We introduce FoundationStereo, a foundation model for stereo depth estimation designed to achieve strong zero-shot generalization. To this end, we first construct a large-scale (1M stereo pairs) synthetic training dataset featuring large diversity and high photorealism, followed by an automatic self-curation pipeline to remove ambiguous samples. We then design a number of network architecture components to enhance scalability, including a side-tuning feature backbone that adapts rich monocular priors from vision foundation models to mitigate the sim-to-real gap, and long-range context reasoning for effective cost volume filtering. Together, these components lead to strong robustness and accuracy across domains, establishing a new standard in zero-shot stereo depth estimation.
♻ ☆ Optimal Virtual Model Control for Robotics: Design and Tuning of Passivity-Based Controllers
Passivity-based control is a cornerstone of control theory and an established design approach in robotics. Its strength is based on the passivity theorem, which provides a powerful interconnection framework for robotics. However, the design of passivity-based controllers and their optimal tuning remain challenging. We propose here an intuitive design approach for fully actuated robots, where the control action is determined by a `virtual-mechanism' as in classical virtual model control. The result is a robot whose controlled behavior can be understood in terms of physics. We achieve optimal tuning by applying algorithmic differentiation to ODE simulations of the rigid body dynamics. Overall, this leads to a flexible design and optimization approach: stability is proven by passivity of the virtual mechanism, while performance is obtained by optimization using algorithmic differentiation.
comment: 14 pages, 17 figures
♻ ☆ Mesh2SLAM in VR: A Fast Geometry-Based SLAM Framework for Rapid Prototyping in Virtual Reality Applications
SLAM is a foundational technique with broad applications in robotics and AR/VR. SLAM simulations evaluate new concepts, but testing on resource-constrained devices, such as VR HMDs, faces challenges: high computational cost and restricted sensor data access. This work proposes a sparse framework using mesh geometry projections as features, which improves efficiency and circumvents direct sensor data access, advancing SLAM research as we demonstrate in VR and through numerical evaluation.
♻ ☆ STPOTR: Simultaneous Human Trajectory and Pose Prediction Using a Non-Autoregressive Transformer for Robot Following Ahead
In this paper, we develop a neural network model to predict future human motion from an observed human motion history. We propose a non-autoregressive transformer architecture to leverage its parallel nature for easier training and fast, accurate predictions at test time. The proposed architecture divides human motion prediction into two parts: 1) the human trajectory, which is the hip joint 3D position over time and 2) the human pose which is the all other joints 3D positions over time with respect to a fixed hip joint. We propose to make the two predictions simultaneously, as the shared representation can improve the model performance. Therefore, the model consists of two sets of encoders and decoders. First, a multi-head attention module applied to encoder outputs improves human trajectory. Second, another multi-head self-attention module applied to encoder outputs concatenated with decoder outputs facilitates learning of temporal dependencies. Our model is well-suited for robotic applications in terms of test accuracy and speed, and compares favorably with respect to state-of-the-art methods. We demonstrate the real-world applicability of our work via the Robot Follow-Ahead task, a challenging yet practical case study for our proposed model.
♻ ☆ ESVO2: Direct Visual-Inertial Odometry with Stereo Event Cameras
Event-based visual odometry is a specific branch of visual Simultaneous Localization and Mapping (SLAM) techniques, which aims at solving tracking and mapping subproblems (typically in parallel), by exploiting the special working principles of neuromorphic (i.e., event-based) cameras. Due to the motion-dependent nature of event data, explicit data association (i.e., feature matching) under large-baseline view-point changes is difficult to establish, making direct methods a more rational choice. However, state-of-the-art direct methods are limited by the high computational complexity of the mapping sub-problem and the degeneracy of camera pose tracking in certain degrees of freedom (DoF) in rotation. In this paper, we tackle these issues by building an event-based stereo visual-inertial odometry system on top of a direct pipeline. Specifically, to speed up the mapping operation, we propose an efficient strategy for sampling contour points according to the local dynamics of events. The mapping performance is also improved in terms of structure completeness and local smoothness by merging the temporal stereo and static stereo results. To circumvent the degeneracy of camera pose tracking in recovering the pitch and yaw components of general 6-DoF motion, we introduce IMU measurements as motion priors via pre-integration. To this end, a compact back-end is proposed for continuously updating the IMU bias and predicting the linear velocity, enabling an accurate motion prediction for camera pose tracking. The resulting system scales well with modern high-resolution event cameras and leads to better global positioning accuracy in large-scale outdoor environments. Extensive evaluations on five publicly available datasets featuring different resolutions and scenarios justify the superior performance of the proposed system against five state-of-the-art methods.
♻ ☆ BILTS: A Bi-Invariant Similarity Measure for Robust Object Trajectory Recognition under Reference Frame Variations
When similar object motions are performed in diverse contexts but are meant to be recognized under a single classification, these contextual variations act as disturbances that negatively affect accurate motion recognition. In this paper, we focus on contextual variations caused by reference frame variations. To robustly deal with these variations, similarity measures have been introduced that compare object motion trajectories in a context-invariant manner. However, most are highly sensitive to noise near singularities, where the measure is not uniquely defined, and lack bi-invariance (invariance to both world and body frame variations). To address these issues, we propose the novel \textit{Bi-Invariant Local Trajectory-Shape Similarity} (BILTS) measure. Compared to other measures, the BILTS measure uniquely offers bi-invariance, boundedness, and third-order shape identity. Aimed at practical implementations, we devised a discretized and regularized version of the BILTS measure which shows exceptional robustness to singularities. This is demonstrated through rigorous recognition experiments using multiple datasets. On average, BILTS attained the highest recognition ratio and least sensitivity to contextual variations compared to other invariant object motion similarity measures. We believe that the BILTS measure is a valuable tool for recognizing motions performed in diverse contexts and has potential in other applications, including the recognition, segmentation, and adaptation of both motion and force trajectories.
comment: This work has been submitted as a regular research paper for consideration in the Journal of Intelligent & Robotic Systems. The content in this preprint is identical to the version submitted for peer review, except for formatting differences required by the journal
♻ ☆ Safe Interval Randomized Path Planning For Manipulators ICAPS 2025
Planning safe paths in 3D workspace for high DoF robotic systems, such as manipulators, is a challenging problem, especially when the environment is populated with the dynamic obstacles that need to be avoided. In this case the time dimension should be taken into account that further increases the complexity of planning. To mitigate this issue we suggest to combine safe-interval path planning (a prominent technique in heuristic search) with the randomized planning, specifically, with the bidirectional rapidly-exploring random trees (RRT-Connect) - a fast and efficient algorithm for high-dimensional planning. Leveraging a dedicated technique of fast computation of the safe intervals we end up with an efficient planner dubbed SI-RRT. We compare it with the state of the art and show that SI-RRT consistently outperforms the competitors both in runtime and solution cost. Our implementation of SI-RRT is publicly available at https://github.com/PathPlanning/ManipulationPlanning-SI-RRT
comment: Submitted to The 35th International Conference on Automated Planning and Scheduling (ICAPS 2025)
♻ ☆ Sensor-Based Distributionally Robust Control for Safe Robot Navigation in Dynamic Environments
We introduce a novel method for mobile robot navigation in dynamic, unknown environments, leveraging onboard sensing and distributionally robust optimization to impose probabilistic safety constraints. Our method introduces a distributionally robust control barrier function (DR-CBF) that directly integrates noisy sensor measurements and state estimates to define safety constraints. This approach is applicable to a wide range of control-affine dynamics, generalizable to robots with complex geometries, and capable of operating at real-time control frequencies. Coupled with a control Lyapunov function (CLF) for path following, the proposed CLF-DR-CBF control synthesis method achieves safe, robust, and efficient navigation in challenging environments. We demonstrate the effectiveness and robustness of our approach for safe autonomous navigation under uncertainty in simulations and real-world experiments with differential-drive robots.
comment: Project page: https://existentialrobotics.org/DRO_Safe_Navigation
♻ ☆ Uncertainty-Aware Planning for Heterogeneous Robot Teams using Dynamic Topological Graphs and Mixed-Integer Programming
Multi-robot planning and coordination in uncertain environments is a fundamental computational challenge, since the belief space increases exponentially with the number of robots. In this paper, we address the problem of planning in uncertain environments with a heterogeneous robot team of fast scout vehicles for information gathering and more risk-averse carrier robots from which the scouts vehicles are deployed. To overcome the computational challenges, we represent the environment and operational scenario using a topological graph, where the parameters of the edge weight distributions vary with the state of the robot team on the graph, and we formulate a computationally efficient mixed-integer program which removes the dependence on the number of robots from its decision space. Our formulation results in the capability to generate optimal multi-robot, long-horizon plans in seconds that could otherwise be computationally intractable. Ultimately our approach enables real-time re-planning, since the computation time is significantly faster than the time to execute one step. We evaluate our approach in a scenario where the robot team must traverse an environment while minimizing detection by observers in positions that are uncertain to the robot team. We demonstrate that our approach is computationally tractable, can improve performance in the presence of imperfect information, and can be adjusted for different risk profiles.
comment: This work has been submitted to the IEEE for possible publication
♻ ☆ A Systematic Study of Multi-Agent Deep Reinforcement Learning for Safe and Robust Autonomous Highway Ramp Entry
Vehicles today can drive themselves on highways and driverless robotaxis operate in major cities, with more sophisticated levels of autonomous driving expected to be available and become more common in the future. Yet, technically speaking, so-called "Level 5" (L5) operation, corresponding to full autonomy, has not been achieved. For that to happen, functions such as fully autonomous highway ramp entry must be available, and provide provably safe, and reliably robust behavior to enable full autonomy. We present a systematic study of a highway ramp function that controls the vehicles forward-moving actions to minimize collisions with the stream of highway traffic into which a merging (ego) vehicle enters. We take a game-theoretic multi-agent (MA) approach to this problem and study the use of controllers based on deep reinforcement learning (DRL). The virtual environment of the MA DRL uses self-play with simulated data where merging vehicles safely learn to control longitudinal position during a taper-type merge. The work presented in this paper extends existing work by studying the interaction of more than two vehicles (agents) and does so by systematically expanding the road scene with additional traffic and ego vehicles. While previous work on the two-vehicle setting established that collision-free controllers are theoretically impossible in fully decentralized, non-coordinated environments, we empirically show that controllers learned using our approach are nearly ideal when measured against idealized optimal controllers.
comment: 9 pages, 9 figures; added support ack
♻ ☆ Beyond Uncertainty: Risk-Aware Active View Acquisition for Safe Robot Navigation and 3D Scene Understanding with FisherRF
The active view acquisition problem has been extensively studied in the context of robot navigation using NeRF and 3D Gaussian Splatting. To enhance scene reconstruction efficiency and ensure robot safety, we propose the Risk-aware Environment Masking (RaEM) framework. RaEM leverages coherent risk measures to dynamically prioritize safety-critical regions of the unknown environment, guiding active view acquisition algorithms toward identifying the next-best-view (NBV). Integrated with FisherRF, which selects the NBV by maximizing expected information gain, our framework achieves a dual objective: improving robot safety and increasing efficiency in risk-aware 3D scene reconstruction and understanding. Extensive high-fidelity experiments validate the effectiveness of our approach, demonstrating its ability to establish a robust and safety-focused framework for active robot exploration and 3D scene understanding.
Multimedia 4
☆ Using Technology in Digital Humanities for Learning and Knowledge Dissemination
Research on Digital Humanities (DH) has been boosted due to the investment in technology for developing access and interaction tools for handling Humanities and Heritage data. The availability of these tools lowers the distance between DH scholars and data generators, and students at various levels, not only because it facilitates access to information but also through the dissemination technologies used in these tools, designed for the improvement of user experience. Most of the disciplines associated with the humanities involve geographical and temporal references, often integrated. These references have been scientifically and pedagogically handled for centuries and are established through the use of maps and timelines. Both these supports have been implemented and used digitally and their potential has been risen through their innovative integration with narratives, storytelling and story maps, enabling the telling of historical events in narratives superimposed on maps. These can be enhanced when supported by rich data, such as images, videos, sound, and their possible combinations in virtual and augmented reality. In this paper, we describe an initial set of tools which use a subset of these technologies and data types to enable learning and dissemination of Humanities data and knowledge. We describe how techniques for making data available and tools for enhancing interaction with these data can improve user experience and potentiate learning and dissemination.
☆ Robust Change Captioning in Remote Sensing: SECOND-CC Dataset and MModalCC Framework
Remote sensing change captioning (RSICC) aims to describe changes between bitemporal images in natural language. Existing methods often fail under challenges like illumination differences, viewpoint changes, blur effects, leading to inaccuracies, especially in no-change regions. Moreover, the images acquired at different spatial resolutions and have registration errors tend to affect the captions. To address these issues, we introduce SECOND-CC, a novel RSICC dataset featuring high-resolution RGB image pairs, semantic segmentation maps, and diverse real-world scenarios. SECOND-CC which contains 6,041 pairs of bitemporal RS images and 30,205 sentences describing the differences between images. Additionally, we propose MModalCC, a multimodal framework that integrates semantic and visual data using advanced attention mechanisms, including Cross-Modal Cross Attention (CMCA) and Multimodal Gated Cross Attention (MGCA). Detailed ablation studies and attention visualizations further demonstrate its effectiveness and ability to address RSICC challenges. Comprehensive experiments show that MModalCC outperforms state-of-the-art RSICC methods, including RSICCformer, Chg2Cap, and PSNet with +4.6% improvement on BLEU4 score and +9.6% improvement on CIDEr score. We will make our dataset and codebase publicly available to facilitate future research at https://github.com/ChangeCapsInRS/SecondCC
comment: This work has been submitted to the IEEE Transactions on Geoscience and Remote Sensing journal for possible publication
☆ CLIP-PCQA: Exploring Subjective-Aligned Vision-Language Modeling for Point Cloud Quality Assessment
In recent years, No-Reference Point Cloud Quality Assessment (NR-PCQA) research has achieved significant progress. However, existing methods mostly seek a direct mapping function from visual data to the Mean Opinion Score (MOS), which is contradictory to the mechanism of practical subjective evaluation. To address this, we propose a novel language-driven PCQA method named CLIP-PCQA. Considering that human beings prefer to describe visual quality using discrete quality descriptions (e.g., "excellent" and "poor") rather than specific scores, we adopt a retrieval-based mapping strategy to simulate the process of subjective assessment. More specifically, based on the philosophy of CLIP, we calculate the cosine similarity between the visual features and multiple textual features corresponding to different quality descriptions, in which process an effective contrastive loss and learnable prompts are introduced to enhance the feature extraction. Meanwhile, given the personal limitations and bias in subjective experiments, we further covert the feature similarities into probabilities and consider the Opinion Score Distribution (OSD) rather than a single MOS as the final target. Experimental results show that our CLIP-PCQA outperforms other State-Of-The-Art (SOTA) approaches.
☆ GVMGen: A General Video-to-Music Generation Model with Hierarchical Attentions AAAI
Composing music for video is essential yet challenging, leading to a growing interest in automating music generation for video applications. Existing approaches often struggle to achieve robust music-video correspondence and generative diversity, primarily due to inadequate feature alignment methods and insufficient datasets. In this study, we present General Video-to-Music Generation model (GVMGen), designed for generating high-related music to the video input. Our model employs hierarchical attentions to extract and align video features with music in both spatial and temporal dimensions, ensuring the preservation of pertinent features while minimizing redundancy. Remarkably, our method is versatile, capable of generating multi-style music from different video inputs, even in zero-shot scenarios. We also propose an evaluation model along with two novel objective metrics for assessing video-music alignment. Additionally, we have compiled a large-scale dataset comprising diverse types of video-music pairs. Experimental results demonstrate that GVMGen surpasses previous models in terms of music-video correspondence, generative diversity, and application universality.
comment: Accepted by the 39th AAAI Conference on Artificial Intelligence (AAAI-25)
Computer Vision and Pattern Recognition 122
☆ Distilling Multi-modal Large Language Models for Autonomous Driving
Autonomous driving demands safe motion planning, especially in critical "long-tail" scenarios. Recent end-to-end autonomous driving systems leverage large language models (LLMs) as planners to improve generalizability to rare events. However, using LLMs at test time introduces high computational costs. To address this, we propose DiMA, an end-to-end autonomous driving system that maintains the efficiency of an LLM-free (or vision-based) planner while leveraging the world knowledge of an LLM. DiMA distills the information from a multi-modal LLM to a vision-based end-to-end planner through a set of specially designed surrogate tasks. Under a joint training strategy, a scene encoder common to both networks produces structured representations that are semantically grounded as well as aligned to the final planning objective. Notably, the LLM is optional at inference, enabling robust planning without compromising on efficiency. Training with DiMA results in a 37% reduction in the L2 trajectory error and an 80% reduction in the collision rate of the vision-based planner, as well as a 44% trajectory error reduction in longtail scenarios. DiMA also achieves state-of-the-art performance on the nuScenes planning benchmark.
☆ SynthLight: Portrait Relighting with Diffusion Model by Learning to Re-render Synthetic Faces
We introduce SynthLight, a diffusion model for portrait relighting. Our approach frames image relighting as a re-rendering problem, where pixels are transformed in response to changes in environmental lighting conditions. Using a physically-based rendering engine, we synthesize a dataset to simulate this lighting-conditioned transformation with 3D head assets under varying lighting. We propose two training and inference strategies to bridge the gap between the synthetic and real image domains: (1) multi-task training that takes advantage of real human portraits without lighting labels; (2) an inference time diffusion sampling procedure based on classifier-free guidance that leverages the input portrait to better preserve details. Our method generalizes to diverse real photographs and produces realistic illumination effects, including specular highlights and cast shadows, while preserving the subject's identity. Our quantitative experiments on Light Stage data demonstrate results comparable to state-of-the-art relighting methods. Our qualitative results on in-the-wild images showcase rich and unprecedented illumination effects. Project Page: \url{https://vrroom.github.io/synthlight/}
comment: 27 pages, 25 figures, Project Page https://vrroom.github.io/synthlight/
☆ Learnings from Scaling Visual Tokenizers for Reconstruction and Generation
Visual tokenization via auto-encoding empowers state-of-the-art image and video generative models by compressing pixels into a latent space. Although scaling Transformer-based generators has been central to recent advances, the tokenizer component itself is rarely scaled, leaving open questions about how auto-encoder design choices influence both its objective of reconstruction and downstream generative performance. Our work aims to conduct an exploration of scaling in auto-encoders to fill in this blank. To facilitate this exploration, we replace the typical convolutional backbone with an enhanced Vision Transformer architecture for Tokenization (ViTok). We train ViTok on large-scale image and video datasets far exceeding ImageNet-1K, removing data constraints on tokenizer scaling. We first study how scaling the auto-encoder bottleneck affects both reconstruction and generation -- and find that while it is highly correlated with reconstruction, its relationship with generation is more complex. We next explored the effect of separately scaling the auto-encoders' encoder and decoder on reconstruction and generation performance. Crucially, we find that scaling the encoder yields minimal gains for either reconstruction or generation, while scaling the decoder boosts reconstruction but the benefits for generation are mixed. Building on our exploration, we design ViTok as a lightweight auto-encoder that achieves competitive performance with state-of-the-art auto-encoders on ImageNet-1K and COCO reconstruction tasks (256p and 512p) while outperforming existing auto-encoders on 16-frame 128p video reconstruction for UCF-101, all with 2-5x fewer FLOPs. When integrated with Diffusion Transformers, ViTok demonstrates competitive performance on image generation for ImageNet-1K and sets new state-of-the-art benchmarks for class-conditional video generation on UCF-101.
comment: 28 pages, 25 figures, 7 Tables
☆ Lost in Translation, Found in Context: Sign Language Translation with Contextual Cues
Our objective is to translate continuous sign language into spoken language text. Inspired by the way human interpreters rely on context for accurate translation, we incorporate additional contextual cues together with the signing video, into a new translation framework. Specifically, besides visual sign recognition features that encode the input video, we integrate complementary textual information from (i) captions describing the background show, (ii) translation of previous sentences, as well as (iii) pseudo-glosses transcribing the signing. These are automatically extracted and inputted along with the visual features to a pre-trained large language model (LLM), which we fine-tune to generate spoken language translations in text form. Through extensive ablation studies, we show the positive contribution of each input cue to the translation performance. We train and evaluate our approach on BOBSL -- the largest British Sign Language dataset currently available. We show that our contextual approach significantly enhances the quality of the translations compared to previously reported results on BOBSL, and also to state-of-the-art methods that we implement as baselines. Furthermore, we demonstrate the generality of our approach by applying it also to How2Sign, an American Sign Language dataset, and achieve competitive results.
☆ SRE-Conv: Symmetric Rotation Equivariant Convolution for Biomedical Image Classification
Convolutional neural networks (CNNs) are essential tools for computer vision tasks, but they lack traditionally desired properties of extracted features that could further improve model performance, e.g., rotational equivariance. Such properties are ubiquitous in biomedical images, which often lack explicit orientation. While current work largely relies on data augmentation or explicit modules to capture orientation information, this comes at the expense of increased training costs or ineffective approximations of the desired equivariance. To overcome these challenges, we propose a novel and efficient implementation of the Symmetric Rotation-Equivariant (SRE) Convolution (SRE-Conv) kernel, designed to learn rotation-invariant features while simultaneously compressing the model size. The SRE-Conv kernel can easily be incorporated into any CNN backbone. We validate the ability of a deep SRE-CNN to capture equivariance to rotation using the public MedMNISTv2 dataset (16 total tasks). SRE-Conv-CNN demonstrated improved rotated image classification performance accuracy on all 16 test datasets in both 2D and 3D images, all while increasing efficiency with fewer parameters and reduced memory footprint. The code is available at https://github.com/XYPB/SRE-Conv.
comment: Accepted by IEEE ISBI 2025 4-page paper
☆ ComplexVAD: Detecting Interaction Anomalies in Video WACV
Existing video anomaly detection datasets are inadequate for representing complex anomalies that occur due to the interactions between objects. The absence of complex anomalies in previous video anomaly detection datasets affects research by shifting the focus onto simple anomalies. To address this problem, we introduce a new large-scale dataset: ComplexVAD. In addition, we propose a novel method to detect complex anomalies via modeling the interactions between objects using a scene graph with spatio-temporal attributes. With our proposed method and two other state-of-the-art video anomaly detection methods, we obtain baseline scores on ComplexVAD and demonstrate that our new method outperforms existing works.
comment: 16 pages, 11 figures, to appear in WACV Workshop ASTAD 2025
☆ Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps
Generative models have made significant impacts across various domains, largely due to their ability to scale during training by increasing data, computational resources, and model size, a phenomenon characterized by the scaling laws. Recent research has begun to explore inference-time scaling behavior in Large Language Models (LLMs), revealing how performance can further improve with additional computation during inference. Unlike LLMs, diffusion models inherently possess the flexibility to adjust inference-time computation via the number of denoising steps, although the performance gains typically flatten after a few dozen. In this work, we explore the inference-time scaling behavior of diffusion models beyond increasing denoising steps and investigate how the generation performance can further improve with increased computation. Specifically, we consider a search problem aimed at identifying better noises for the diffusion sampling process. We structure the design space along two axes: the verifiers used to provide feedback, and the algorithms used to find better noise candidates. Through extensive experiments on class-conditioned and text-conditioned image generation benchmarks, our findings reveal that increasing inference-time compute leads to substantial improvements in the quality of samples generated by diffusion models, and with the complicated nature of images, combinations of the components in the framework can be specifically chosen to conform with different application scenario.
☆ A Simple Aerial Detection Baseline of Multimodal Language Models
The multimodal language models (MLMs) based on generative pre-trained Transformer are considered powerful candidates for unifying various domains and tasks. MLMs developed for remote sensing (RS) have demonstrated outstanding performance in multiple tasks, such as visual question answering and visual grounding. In addition to visual grounding that detects specific objects corresponded to given instruction, aerial detection, which detects all objects of multiple categories, is also a valuable and challenging task for RS foundation models. However, aerial detection has not been explored by existing RS MLMs because the autoregressive prediction mechanism of MLMs differs significantly from the detection outputs. In this paper, we present a simple baseline for applying MLMs to aerial detection for the first time, named LMMRotate. Specifically, we first introduce a normalization method to transform detection outputs into textual outputs to be compatible with the MLM framework. Then, we propose a evaluation method, which ensures a fair comparison between MLMs and conventional object detection models. We construct the baseline by fine-tuning open-source general-purpose MLMs and achieve impressive detection performance comparable to conventional detector. We hope that this baseline will serve as a reference for future MLM development, enabling more comprehensive capabilities for understanding RS images. Code is available at https://github.com/Li-Qingyun/mllm-mmrotate.
comment: 4 pages, 1 table, 4 figures
☆ FLOL: Fast Baselines for Real-World Low-Light Enhancement
Low-Light Image Enhancement (LLIE) is a key task in computational photography and imaging. The problem of enhancing images captured during night or in dark environments has been well-studied in the image signal processing literature. However, current deep learning-based solutions struggle with efficiency and robustness in real-world scenarios (e.g. scenes with noise, saturated pixels, bad illumination). We propose a lightweight neural network that combines image processing in the frequency and spatial domains. Our method, FLOL+, is one of the fastest models for this task, achieving state-of-the-art results on popular real scenes datasets such as LOL and LSRW. Moreover, we are able to process 1080p images under 12ms. Code and models at https://github.com/cidautai/FLOL
comment: Technical Report
☆ Practical Continual Forgetting for Pre-trained Vision Models
For privacy and security concerns, the need to erase unwanted information from pre-trained vision models is becoming evident nowadays. In real-world scenarios, erasure requests originate at any time from both users and model owners, and these requests usually form a sequence. Therefore, under such a setting, selective information is expected to be continuously removed from a pre-trained model while maintaining the rest. We define this problem as continual forgetting and identify three key challenges. (i) For unwanted knowledge, efficient and effective deleting is crucial. (ii) For remaining knowledge, the impact brought by the forgetting procedure should be minimal. (iii) In real-world scenarios, the training samples may be scarce or partially missing during the process of forgetting. To address them, we first propose Group Sparse LoRA (GS-LoRA). Specifically, towards (i), we introduce LoRA modules to fine-tune the FFN layers in Transformer blocks for each forgetting task independently, and towards (ii), a simple group sparse regularization is adopted, enabling automatic selection of specific LoRA groups and zeroing out the others. To further extend GS-LoRA to more practical scenarios, we incorporate prototype information as additional supervision and introduce a more practical approach, GS-LoRA++. For each forgotten class, we move the logits away from its original prototype. For the remaining classes, we pull the logits closer to their respective prototypes. We conduct extensive experiments on face recognition, object detection and image classification and demonstrate that our method manages to forget specific classes with minimal impact on other classes. Codes have been released on https://github.com/bjzhb666/GS-LoRA.
☆ Mitigating Hallucinations in Large Vision-Language Models via DPO: On-Policy Data Hold the Key
Hallucination remains a major challenge for Large Vision-Language Models (LVLMs). Direct Preference Optimization (DPO) has gained increasing attention as a simple solution to hallucination issues. It directly learns from constructed preference pairs that reflect the severity of hallucinations in responses to the same prompt and image. Nonetheless, different data construction methods in existing works bring notable performance variations. We identify a crucial factor here: outcomes are largely contingent on whether the constructed data aligns on-policy w.r.t the initial (reference) policy of DPO. Theoretical analysis suggests that learning from off-policy data is impeded by the presence of KL-divergence between the updated policy and the reference policy. From the perspective of dataset distribution, we systematically summarize the inherent flaws in existing algorithms that employ DPO to address hallucination issues. To alleviate the problems, we propose On-Policy Alignment (OPA)-DPO framework, which uniquely leverages expert feedback to correct hallucinated responses and aligns both the original and expert-revised responses in an on-policy manner. Notably, with only 4.8k data, OPA-DPO achieves an additional reduction in the hallucination rate of LLaVA-1.5-7B: 13.26% on the AMBER benchmark and 5.39% on the Object-Hal benchmark, compared to the previous SOTA algorithm trained with 16k samples.
comment: 18 pages, 15 figures
☆ Fine-Grained Image-Text Correspondence with Cost Aggregation for Open-Vocabulary Part Segmentation
Open-Vocabulary Part Segmentation (OVPS) is an emerging field for recognizing fine-grained parts in unseen categories. We identify two primary challenges in OVPS: (1) the difficulty in aligning part-level image-text correspondence, and (2) the lack of structural understanding in segmenting object parts. To address these issues, we propose PartCATSeg, a novel framework that integrates object-aware part-level cost aggregation, compositional loss, and structural guidance from DINO. Our approach employs a disentangled cost aggregation strategy that handles object and part-level costs separately, enhancing the precision of part-level segmentation. We also introduce a compositional loss to better capture part-object relationships, compensating for the limited part annotations. Additionally, structural guidance from DINO features improves boundary delineation and inter-part understanding. Extensive experiments on Pascal-Part-116, ADE20K-Part-234, and PartImageNet datasets demonstrate that our method significantly outperforms state-of-the-art approaches, setting a new baseline for robust generalization to unseen part categories.
☆ Robin: a Suite of Multi-Scale Vision-Language Models and the CHIRP Evaluation Benchmark
The proliferation of Vision-Language Models (VLMs) in the past several years calls for rigorous and comprehensive evaluation methods and benchmarks. This work analyzes existing VLM evaluation techniques, including automated metrics, AI-based assessments, and human evaluations across diverse tasks. We first introduce Robin - a novel suite of VLMs that we built by combining Large Language Models (LLMs) and Vision Encoders (VEs) at multiple scales, and use Robin to identify shortcomings of current evaluation approaches across scales. Next, to overcome the identified limitations, we introduce CHIRP - a new long form response benchmark we developed for more robust and complete VLM evaluation. We provide open access to the Robin training code, model suite, and CHIRP benchmark to promote reproducibility and advance VLM research.
☆ Unified Face Matching and Physical-Digital Spoofing Attack Detection
Face recognition technology has dramatically transformed the landscape of security, surveillance, and authentication systems, offering a user-friendly and non-invasive biometric solution. However, despite its significant advantages, face recognition systems face increasing threats from physical and digital spoofing attacks. Current research typically treats face recognition and attack detection as distinct classification challenges. This approach necessitates the implementation of separate models for each task, leading to considerable computational complexity, particularly on devices with limited resources. Such inefficiencies can stifle scalability and hinder performance. In response to these challenges, this paper introduces an innovative unified model designed for face recognition and detection of physical and digital attacks. By leveraging the advanced Swin Transformer backbone and incorporating HiLo attention in a convolutional neural network framework, we address unified face recognition and spoof attack detection more effectively. Moreover, we introduce augmentation techniques that replicate the traits of physical and digital spoofing cues, significantly enhancing our model robustness. Through comprehensive experimental evaluation across various datasets, we showcase the effectiveness of our model in unified face recognition and spoof detection. Additionally, we confirm its resilience against unseen physical and digital spoofing attacks, underscoring its potential for real-world applications.
☆ WMamba: Wavelet-based Mamba for Face Forgery Detection
With the rapid advancement of deepfake generation technologies, the demand for robust and accurate face forgery detection algorithms has become increasingly critical. Recent studies have demonstrated that wavelet analysis can uncover subtle forgery artifacts that remain imperceptible in the spatial domain. Wavelets effectively capture important facial contours, which are often slender, fine-grained, and global in nature. However, existing wavelet-based approaches fail to fully leverage these unique characteristics, resulting in sub-optimal feature extraction and limited generalizability. To address this challenge, we introduce WMamba, a novel wavelet-based feature extractor built upon the Mamba architecture. WMamba maximizes the utility of wavelet information through two key innovations. First, we propose Dynamic Contour Convolution (DCConv), which employs specially crafted deformable kernels to adaptively model slender facial contours. Second, by leveraging the Mamba architecture, our method captures long-range spatial relationships with linear computational complexity. This efficiency allows for the extraction of fine-grained, global forgery artifacts from small image patches. Extensive experimental results show that WMamba achieves state-of-the-art (SOTA) performance, highlighting its effectiveness and superiority in face forgery detection.
☆ Metric Learning with Progressive Self-Distillation for Audio-Visual Embedding Learning ICASSP 2025
Metric learning projects samples into an embedded space, where similarities and dissimilarities are quantified based on their learned representations. However, existing methods often rely on label-guided representation learning, where representations of different modalities, such as audio and visual data, are aligned based on annotated labels. This approach tends to underutilize latent complex features and potential relationships inherent in the distributions of audio and visual data that are not directly tied to the labels, resulting in suboptimal performance in audio-visual embedding learning. To address this issue, we propose a novel architecture that integrates cross-modal triplet loss with progressive self-distillation. Our method enhances representation learning by leveraging inherent distributions and dynamically refining soft audio-visual alignments -- probabilistic alignments between audio and visual data that capture the inherent relationships beyond explicit labels. Specifically, the model distills audio-visual distribution-based knowledge from annotated labels in a subset of each batch. This self-distilled knowledge is used t
comment: 5 pages, 3 figures, 2 tables. Accepted by ICASSP 2025
☆ Mesh2SLAM in VR: A Fast Geometry-Based SLAM Framework for Rapid Prototyping in Virtual Reality Applications
SLAM is a foundational technique with broad applications in robotics and AR/VR. SLAM simulations evaluate new concepts, but testing on resource-constrained devices, such as VR HMDs, faces challenges: high computational cost and restricted sensor data access. This work proposes a sparse framework using mesh geometry projections as features, which improves efficiency and circumvents direct sensor data access, advancing SLAM research as we demonstrate in VR and through numerical evaluation.
☆ Sequential PatchCore: Anomaly Detection for Surface Inspection using Synthetic Impurities
The appearance of surface impurities (e.g., water stains, fingerprints, stickers) is an often-mentioned issue that causes degradation of automated visual inspection systems. At the same time, synthetic data generation techniques for visual surface inspection have focused primarily on generating perfect examples and defects, disregarding impurities. This study highlights the importance of considering impurities when generating synthetic data. We introduce a procedural method to include photorealistic water stains in synthetic data. The synthetic datasets are generated to correspond to real datasets and are further used to train an anomaly detection model and investigate the influence of water stains. The high-resolution images used for surface inspection lead to memory bottlenecks during anomaly detection training. To address this, we introduce Sequential PatchCore - a method to build coresets sequentially and make training on large images using consumer-grade hardware tractable. This allows us to perform transfer learning using coresets pre-trained on different dataset versions. Our results show the benefits of using synthetic data for pre-training an explicit coreset anomaly model and the extended performance benefits of finetuning the coreset using real data. We observed how the impurities and labelling ambiguity lower the model performance and have additionally reported the defect-wise recall to provide an industrially relevant perspective on model performance.
☆ A New Teacher-Reviewer-Student Framework for Semi-supervised 2D Human Pose Estimation
Conventional 2D human pose estimation methods typically require extensive labeled annotations, which are both labor-intensive and expensive. In contrast, semi-supervised 2D human pose estimation can alleviate the above problems by leveraging a large amount of unlabeled data along with a small portion of labeled data. Existing semi-supervised 2D human pose estimation methods update the network through backpropagation, ignoring crucial historical information from the previous training process. Therefore, we propose a novel semi-supervised 2D human pose estimation method by utilizing a newly designed Teacher-Reviewer-Student framework. Specifically, we first mimic the phenomenon that human beings constantly review previous knowledge for consolidation to design our framework, in which the teacher predicts results to guide the student's learning and the reviewer stores important historical parameters to provide additional supervision signals. Secondly, we introduce a Multi-level Feature Learning strategy, which utilizes the outputs from different stages of the backbone to estimate the heatmap to guide network training, enriching the supervisory information while effectively capturing keypoint relationships. Finally, we design a data augmentation strategy, i.e., Keypoint-Mix, to perturb pose information by mixing different keypoints, thus enhancing the network's ability to discern keypoints. Extensive experiments on publicly available datasets, demonstrate our method achieves significant improvements compared to the existing methods.
☆ Text-driven Adaptation of Foundation Models for Few-shot Surgical Workflow Analysis
Purpose: Surgical workflow analysis is crucial for improving surgical efficiency and safety. However, previous studies rely heavily on large-scale annotated datasets, posing challenges in cost, scalability, and reliance on expert annotations. To address this, we propose Surg-FTDA (Few-shot Text-driven Adaptation), designed to handle various surgical workflow analysis tasks with minimal paired image-label data. Methods: Our approach has two key components. First, Few-shot selection-based modality alignment selects a small subset of images and aligns their embeddings with text embeddings from the downstream task, bridging the modality gap. Second, Text-driven adaptation leverages only text data to train a decoder, eliminating the need for paired image-text data. This decoder is then applied to aligned image embeddings, enabling image-related tasks without explicit image-text pairs. Results: We evaluate our approach to generative tasks (image captioning) and discriminative tasks (triplet recognition and phase recognition). Results show that Surg-FTDA outperforms baselines and generalizes well across downstream tasks. Conclusion: We propose a text-driven adaptation approach that mitigates the modality gap and handles multiple downstream tasks in surgical workflow analysis, with minimal reliance on large annotated datasets. The code and dataset will be released in https://github.com/TingxuanSix/Surg-FTDA.
☆ Exploring AI-based System Design for Pixel-level Protected Health Information Detection in Medical Images
De-identification of medical images is a critical step to ensure privacy during data sharing in research and clinical settings. The initial step in this process involves detecting Protected Health Information (PHI), which can be found in image metadata or imprinted within image pixels. Despite the importance of such systems, there has been limited evaluation of existing AI-based solutions, creating barriers to the development of reliable and robust tools. In this study, we present an AI-based pipeline for PHI detection, comprising three key components: text detection, text extraction, and analysis of PHI content in medical images. By experimenting with exchanging roles of vision and language models within the pipeline, we evaluate the performance and recommend the best setup for the PHI detection task.
comment: In progress
☆ AdaFV: Accelerating VLMs with Self-Adaptive Cross-Modality Attention Mixture
The success of VLMs often relies on the dynamic high-resolution schema that adaptively augments the input images to multiple crops, so that the details of the images can be retained. However, such approaches result in a large number of redundant visual tokens, thus significantly reducing the efficiency of the VLMs. To improve the VLMs' efficiency without introducing extra training costs, many research works are proposed to reduce the visual tokens by filtering the uninformative visual tokens or aggregating their information. Some approaches propose to reduce the visual tokens according to the self-attention of VLMs, which are biased, to result in inaccurate responses. The token reduction approaches solely rely on visual cues are text-agnostic, and fail to focus on the areas that are most relevant to the question, especially when the queried objects are non-salient to the image. In this work, we first conduct experiments to show that the original text embeddings are aligned with the visual tokens, without bias on the tailed visual tokens. We then propose a self-adaptive cross-modality attention mixture mechanism that dynamically leverages the effectiveness of visual saliency and text-to-image similarity in the pre-LLM layers to select the visual tokens that are informative. Extensive experiments demonstrate that the proposed approach achieves state-of-the-art training-free VLM acceleration performance, especially when the reduction rate is sufficiently large.
comment: 12 pages, 6 figures
☆ HydraMix: Multi-Image Feature Mixing for Small Data Image Classification
Training deep neural networks requires datasets with a large number of annotated examples. The collection and annotation of these datasets is not only extremely expensive but also faces legal and privacy problems. These factors are a significant limitation for many real-world applications. To address this, we introduce HydraMix, a novel architecture that generates new image compositions by mixing multiple different images from the same class. HydraMix learns the fusion of the content of various images guided by a segmentation-based mixing mask in feature space and is optimized via a combination of unsupervised and adversarial training. Our data augmentation scheme allows the creation of models trained from scratch on very small datasets. We conduct extensive experiments on ciFAIR-10, STL-10, and ciFAIR-100. Additionally, we introduce a novel text-image metric to assess the generality of the augmented datasets. Our results show that HydraMix outperforms existing state-of-the-art methods for image classification on small datasets.
☆ AnyStory: Towards Unified Single and Multiple Subject Personalization in Text-to-Image Generation
Recently, large-scale generative models have demonstrated outstanding text-to-image generation capabilities. However, generating high-fidelity personalized images with specific subjects still presents challenges, especially in cases involving multiple subjects. In this paper, we propose AnyStory, a unified approach for personalized subject generation. AnyStory not only achieves high-fidelity personalization for single subjects, but also for multiple subjects, without sacrificing subject fidelity. Specifically, AnyStory models the subject personalization problem in an "encode-then-route" manner. In the encoding step, AnyStory utilizes a universal and powerful image encoder, i.e., ReferenceNet, in conjunction with CLIP vision encoder to achieve high-fidelity encoding of subject features. In the routing step, AnyStory utilizes a decoupled instance-aware subject router to accurately perceive and predict the potential location of the corresponding subject in the latent space, and guide the injection of subject conditions. Detailed experimental results demonstrate the excellent performance of our method in retaining subject details, aligning text descriptions, and personalizing for multiple subjects. The project page is at https://aigcdesigngroup.github.io/AnyStory/ .
comment: Tech report; Project page: https://aigcdesigngroup.github.io/AnyStory/
☆ Omni-Emotion: Extending Video MLLM with Detailed Face and Audio Modeling for Multimodal Emotion Analysis
Understanding emotions accurately is essential for fields like human-computer interaction. Due to the complexity of emotions and their multi-modal nature (e.g., emotions are influenced by facial expressions and audio), researchers have turned to using multi-modal models to understand human emotions rather than single-modality. However, current video multi-modal large language models (MLLMs) encounter difficulties in effectively integrating audio and identifying subtle facial micro-expressions. Furthermore, the lack of detailed emotion analysis datasets also limits the development of multimodal emotion analysis. To address these issues, we introduce a self-reviewed dataset and a human-reviewed dataset, comprising 24,137 coarse-grained samples and 3,500 manually annotated samples with detailed emotion annotations, respectively. These datasets allow models to learn from diverse scenarios and better generalize to real-world applications. Moreover, in addition to the audio modeling, we propose to explicitly integrate facial encoding models into the existing advanced Video MLLM, enabling the MLLM to effectively unify audio and the subtle facial cues for emotion understanding. By aligning these features within a unified space and employing instruction tuning in our proposed datasets, our Omni-Emotion achieves state-of-the-art performance in both emotion recognition and reasoning tasks.
☆ VanGogh: A Unified Multimodal Diffusion-based Framework for Video Colorization
Video colorization aims to transform grayscale videos into vivid color representations while maintaining temporal consistency and structural integrity. Existing video colorization methods often suffer from color bleeding and lack comprehensive control, particularly under complex motion or diverse semantic cues. To this end, we introduce VanGogh, a unified multimodal diffusion-based framework for video colorization. VanGogh tackles these challenges using a Dual Qformer to align and fuse features from multiple modalities, complemented by a depth-guided generation process and an optical flow loss, which help reduce color overflow. Additionally, a color injection strategy and luma channel replacement are implemented to improve generalization and mitigate flickering artifacts. Thanks to this design, users can exercise both global and local control over the generation process, resulting in higher-quality colorized videos. Extensive qualitative and quantitative evaluations, and user studies, demonstrate that VanGogh achieves superior temporal consistency and color fidelity.Project page: https://becauseimbatman0.github.io/VanGogh.
☆ Comparison of Various SLAM Systems for Mobile Robot in an Indoor Environment
This article presents a comparative analysis of a mobile robot trajectories computed by various ROS-based SLAM systems. For this reason we developed a prototype of a mobile robot with common sensors: 2D lidar, a monocular and ZED stereo cameras. Then we conducted experiments in a typical office environment and collected data from all sensors, running all tested SLAM systems based on the acquired dataset. We studied the following SLAM systems: (a) 2D lidar-based: GMapping, Hector SLAM, Cartographer; (b) monocular camera-based: Large Scale Direct monocular SLAM (LSD SLAM), ORB SLAM, Direct Sparse Odometry (DSO); and (c) stereo camera-based: ZEDfu, Real-Time Appearance-Based Mapping (RTAB map), ORB SLAM, Stereo Parallel Tracking and Mapping (S-PTAM). Since all SLAM methods were tested on the same dataset we compared results for different SLAM systems with appropriate metrics, demonstrating encouraging results for lidar-based Cartographer SLAM, Monocular ORB SLAM and Stereo RTAB Map methods.
comment: 6 pages, 6 figures
☆ The Devil is in the Details: Simple Remedies for Image-to-LiDAR Representation Learning ACCV2024
LiDAR is a crucial sensor in autonomous driving, commonly used alongside cameras. By exploiting this camera-LiDAR setup and recent advances in image representation learning, prior studies have shown the promising potential of image-to-LiDAR distillation. These prior arts focus on the designs of their own losses to effectively distill the pre-trained 2D image representations into a 3D model. However, the other parts of the designs have been surprisingly unexplored. We find that fundamental design elements, e.g., the LiDAR coordinate system, quantization according to the existing input interface, and data utilization, are more critical than developing loss functions, which have been overlooked in prior works. In this work, we show that simple fixes to these designs notably outperform existing methods by 16% in 3D semantic segmentation on the nuScenes dataset and 13% in 3D object detection on the KITTI dataset in downstream task performance. We focus on overlooked design choices along the spatial and temporal axes. Spatially, prior work has used cylindrical coordinate and voxel sizes without considering their side effects yielded with a commonly deployed sparse convolution layer input interface, leading to spatial quantization errors in 3D models. Temporally, existing work has avoided cumbersome data curation by discarding unsynced data, limiting the use to only the small portion of data that is temporally synced across sensors. We analyze these effects and propose simple solutions for each overlooked aspect.
comment: Accepted to ACCV2024
☆ MonoSOWA: Scalable monocular 3D Object detector Without human Annotations
Detecting the three-dimensional position and orientation of objects using a single RGB camera is a foundational task in computer vision with many important applications. Traditionally, 3D object detection methods are trained in a fully-supervised setup, requiring vast amounts of human annotations, which are laborious, costly, and do not scale well with the ever-increasing amounts of data being captured. In this paper, we present the first method to train 3D object detectors for monocular RGB cameras without domain-specific human annotations, thus making orders of magnitude more data available for training. Thanks to newly proposed Canonical Object Space, the method can not only exploit data across a variety of datasets and camera setups to train a single 3D detector, but unlike previous work it also works out of the box in previously unseen camera setups. All this is crucial for practical applications, where the data and cameras are extremely heterogeneous. The method is evaluated on two standard autonomous driving datasets, where it outperforms previous works, which, unlike our method, still rely on 2D human annotations.
☆ DEFOM-Stereo: Depth Foundation Model Based Stereo Matching
Stereo matching is a key technique for metric depth estimation in computer vision and robotics. Real-world challenges like occlusion and non-texture hinder accurate disparity estimation from binocular matching cues. Recently, monocular relative depth estimation has shown remarkable generalization using vision foundation models. Thus, to facilitate robust stereo matching with monocular depth cues, we incorporate a robust monocular relative depth model into the recurrent stereo-matching framework, building a new framework for depth foundation model-based stereo-matching, DEFOM-Stereo. In the feature extraction stage, we construct the combined context and matching feature encoder by integrating features from conventional CNNs and DEFOM. In the update stage, we use the depth predicted by DEFOM to initialize the recurrent disparity and introduce a scale update module to refine the disparity at the correct scale. DEFOM-Stereo is verified to have comparable performance on the Scene Flow dataset with state-of-the-art (SOTA) methods and notably shows much stronger zero-shot generalization. Moreover, DEFOM-Stereo achieves SOTA performance on the KITTI 2012, KITTI 2015, Middlebury, and ETH3D benchmarks, ranking 1st on many metrics. In the joint evaluation under the robust vision challenge, our model simultaneously outperforms previous models on the individual benchmarks. Both results demonstrate the outstanding capabilities of the proposed model.
comment: Code: https://github.com/Insta360-Research-Team/DEFOM-Stereo
☆ RE-POSE: Synergizing Reinforcement Learning-Based Partitioning and Offloading for Edge Object Detection
Object detection plays a crucial role in smart video analysis, with applications ranging from autonomous driving and security to smart cities. However, achieving real-time object detection on edge devices presents significant challenges due to their limited computational resources and the high demands of deep neural network (DNN)-based detection models, particularly when processing high-resolution video. Conventional strategies, such as input down-sampling and network up-scaling, often compromise detection accuracy for faster performance or lead to higher inference latency. To address these issues, this paper introduces RE-POSE, a Reinforcement Learning (RL)-Driven Partitioning and Edge Offloading framework designed to optimize the accuracy-latency trade-off in resource-constrained edge environments. Our approach features an RL-Based Dynamic Clustering Algorithm (RL-DCA) that partitions video frames into non-uniform blocks based on object distribution and the computational characteristics of DNNs. Furthermore, a parallel edge offloading scheme is implemented to distribute these blocks across multiple edge servers for concurrent processing. Experimental evaluations show that RE-POSE significantly enhances detection accuracy and reduces inference latency, surpassing existing methods.
☆ Normal-NeRF: Ambiguity-Robust Normal Estimation for Highly Reflective Scenes AAAI 2025
Neural Radiance Fields (NeRF) often struggle with reconstructing and rendering highly reflective scenes. Recent advancements have developed various reflection-aware appearance models to enhance NeRF's capability to render specular reflections. However, the robust reconstruction of highly reflective scenes is still hindered by the inherent shape ambiguity on specular surfaces. Existing methods typically rely on additional geometry priors to regularize the shape prediction, but this can lead to oversmoothed geometry in complex scenes. Observing the critical role of surface normals in parameterizing reflections, we introduce a transmittance-gradient-based normal estimation technique that remains robust even under ambiguous shape conditions. Furthermore, we propose a dual activated densities module that effectively bridges the gap between smooth surface normals and sharp object boundaries. Combined with a reflection-aware appearance model, our proposed method achieves robust reconstruction and high-fidelity rendering of scenes featuring both highly specular reflections and intricate geometric structures. Extensive experiments demonstrate that our method outperforms existing state-of-the-art methods on various datasets.
comment: AAAI 2025, code available at https://github.com/sjj118/Normal-NeRF
☆ On the Relation between Optical Aperture and Automotive Object Detection
We explore the impact of aperture size and shape on automotive camera systems for deep-learning-based tasks like traffic sign recognition and light state detection. A method is proposed to simulate optical effects using the point spread function (PSF), enhancing realism and reducing the domain gap between synthetic and real-world images. Computer-generated scenes are refined with this technique to model optical distortions and improve simulation accuracy.
☆ Double Visual Defense: Adversarial Pre-training and Instruction Tuning for Improving Vision-Language Model Robustness
This paper investigates the robustness of vision-language models against adversarial visual perturbations and introduces a novel ``double visual defense" to enhance this robustness. Unlike previous approaches that resort to lightweight adversarial fine-tuning of a pre-trained CLIP model, we perform large-scale adversarial vision-language pre-training from scratch using web-scale data. We then strengthen the defense by incorporating adversarial visual instruction tuning. The resulting models from each stage, $\Delta$CLIP and $\Delta^2$LLaVA, show substantially enhanced zero-shot robustness and set a new state-of-the-art in adversarial defense for vision-language models. For example, the adversarial robustness of $\Delta$CLIP surpasses that of the previous best models on ImageNet-1k by ~20%. %For example, $\Delta$CLIP surpasses the previous best models on ImageNet-1k by ~20% in terms of adversarial robustness. Similarly, compared to prior art, $\Delta^2$LLaVA brings a ~30% robustness improvement to image captioning task and a ~20% robustness improvement to visual question answering task. Furthermore, our models exhibit stronger zero-shot recognition capability, fewer hallucinations, and superior reasoning performance compared to baselines. Our project page is https://doublevisualdefense.github.io/.
☆ Scaling up self-supervised learning for improved surgical foundation models
Foundation models have revolutionized computer vision by achieving vastly superior performance across diverse tasks through large-scale pretraining on extensive datasets. However, their application in surgical computer vision has been limited. This study addresses this gap by introducing SurgeNetXL, a novel surgical foundation model that sets a new benchmark in surgical computer vision. Trained on the largest reported surgical dataset to date, comprising over 4.7 million video frames, SurgeNetXL achieves consistent top-tier performance across six datasets spanning four surgical procedures and three tasks, including semantic segmentation, phase recognition, and critical view of safety (CVS) classification. Compared with the best-performing surgical foundation models, SurgeNetXL shows mean improvements of 2.4, 9.0, and 12.6 percent for semantic segmentation, phase recognition, and CVS classification, respectively. Additionally, SurgeNetXL outperforms the best-performing ImageNet-based variants by 14.4, 4.0, and 1.6 percent in the respective tasks. In addition to advancing model performance, this study provides key insights into scaling pretraining datasets, extending training durations, and optimizing model architectures specifically for surgical computer vision. These findings pave the way for improved generalizability and robustness in data-scarce scenarios, offering a comprehensive framework for future research in this domain. All models and a subset of the SurgeNetXL dataset, including over 2 million video frames, are publicly available at: https://github.com/TimJaspers0801/SurgeNet.
☆ CaPa: Carve-n-Paint Synthesis for Efficient 4K Textured Mesh Generation
The synthesis of high-quality 3D assets from textual or visual inputs has become a central objective in modern generative modeling. Despite the proliferation of 3D generation algorithms, they frequently grapple with challenges such as multi-view inconsistency, slow generation times, low fidelity, and surface reconstruction problems. While some studies have addressed some of these issues, a comprehensive solution remains elusive. In this paper, we introduce \textbf{CaPa}, a carve-and-paint framework that generates high-fidelity 3D assets efficiently. CaPa employs a two-stage process, decoupling geometry generation from texture synthesis. Initially, a 3D latent diffusion model generates geometry guided by multi-view inputs, ensuring structural consistency across perspectives. Subsequently, leveraging a novel, model-agnostic Spatially Decoupled Attention, the framework synthesizes high-resolution textures (up to 4K) for a given geometry. Furthermore, we propose a 3D-aware occlusion inpainting algorithm that fills untextured regions, resulting in cohesive results across the entire model. This pipeline generates high-quality 3D assets in less than 30 seconds, providing ready-to-use outputs for commercial applications. Experimental results demonstrate that CaPa excels in both texture fidelity and geometric stability, establishing a new standard for practical, scalable 3D asset generation.
comment: project page: https://ncsoft.github.io/CaPa/
☆ AugRefer: Advancing 3D Visual Grounding via Cross-Modal Augmentation and Spatial Relation-based Referring AAAI 2025
3D visual grounding (3DVG), which aims to correlate a natural language description with the target object within a 3D scene, is a significant yet challenging task. Despite recent advancements in this domain, existing approaches commonly encounter a shortage: a limited amount and diversity of text3D pairs available for training. Moreover, they fall short in effectively leveraging different contextual clues (e.g., rich spatial relations within the 3D visual space) for grounding. To address these limitations, we propose AugRefer, a novel approach for advancing 3D visual grounding. AugRefer introduces cross-modal augmentation designed to extensively generate diverse text-3D pairs by placing objects into 3D scenes and creating accurate and semantically rich descriptions using foundation models. Notably, the resulting pairs can be utilized by any existing 3DVG methods for enriching their training data. Additionally, AugRefer presents a language-spatial adaptive decoder that effectively adapts the potential referring objects based on the language description and various 3D spatial relations. Extensive experiments on three benchmark datasets clearly validate the effectiveness of AugRefer.
comment: AAAI 2025
☆ Vision-Language Models Do Not Understand Negation
Many practical vision-language applications require models that understand negation, e.g., when using natural language to retrieve images which contain certain objects but not others. Despite advancements in vision-language models (VLMs) through large-scale training, their ability to comprehend negation remains underexplored. This study addresses the question: how well do current VLMs understand negation? We introduce NegBench, a new benchmark designed to evaluate negation understanding across 18 task variations and 79k examples spanning image, video, and medical datasets. The benchmark consists of two core tasks designed to evaluate negation understanding in diverse multimodal settings: Retrieval with Negation and Multiple Choice Questions with Negated Captions. Our evaluation reveals that modern VLMs struggle significantly with negation, often performing at chance level. To address these shortcomings, we explore a data-centric approach wherein we finetune CLIP models on large-scale synthetic datasets containing millions of negated captions. We show that this approach can result in a 10% increase in recall on negated queries and a 40% boost in accuracy on multiple-choice questions with negated captions.
comment: Project page: https://negbench.github.io
☆ Dynamic Neural Style Transfer for Artistic Image Generation using VGG19
Throughout history, humans have created remarkable works of art, but artificial intelligence has only recently started to make strides in generating visually compelling art. Breakthroughs in the past few years have focused on using convolutional neural networks (CNNs) to separate and manipulate the content and style of images, applying texture synthesis techniques. Nevertheless, a number of current techniques continue to encounter obstacles, including lengthy processing times, restricted choices of style images, and the inability to modify the weight ratio of styles. We proposed a neural style transfer system that can add various artistic styles to a desired image to address these constraints allowing flexible adjustments to style weight ratios and reducing processing time. The system uses the VGG19 model for feature extraction, ensuring high-quality, flexible stylization without compromising content integrity.
☆ Towards Robust and Realistic Human Pose Estimation via WiFi Signals
Robust WiFi-based human pose estimation is a challenging task that bridges discrete and subtle WiFi signals to human skeletons. This paper revisits this problem and reveals two critical yet overlooked issues: 1) cross-domain gap, i.e., due to significant variations between source-target domain pose distributions; and 2) structural fidelity gap, i.e., predicted skeletal poses manifest distorted topology, usually with misplaced joints and disproportionate bone lengths. This paper fills these gaps by reformulating the task into a novel two-phase framework dubbed DT-Pose: Domain-consistent representation learning and Topology-constrained Pose decoding. Concretely, we first propose a temporal-consistent contrastive learning strategy with uniformity regularization, coupled with self-supervised masking-reconstruction operations, to enable robust learning of domain-consistent and motion-discriminative WiFi-specific representations. Beyond this, we introduce a simple yet effective pose decoder with task prompts, which integrates Graph Convolution Network (GCN) and Transformer layers to constrain the topology structure of the generated skeleton by exploring the adjacent-overarching relationships among human joints. Extensive experiments conducted on various benchmark datasets highlight the superior performance of our method in tackling these fundamental challenges in both 2D/3D human pose estimation tasks.
comment: 15 pages, 9 figures
☆ PISCO: Self-Supervised k-Space Regularization for Improved Neural Implicit k-Space Representations of Dynamic MRI
Neural implicit k-space representations (NIK) have shown promising results for dynamic magnetic resonance imaging (MRI) at high temporal resolutions. Yet, reducing acquisition time, and thereby available training data, results in severe performance drops due to overfitting. To address this, we introduce a novel self-supervised k-space loss function $\mathcal{L}_\mathrm{PISCO}$, applicable for regularization of NIK-based reconstructions. The proposed loss function is based on the concept of parallel imaging-inspired self-consistency (PISCO), enforcing a consistent global k-space neighborhood relationship without requiring additional data. Quantitative and qualitative evaluations on static and dynamic MR reconstructions show that integrating PISCO significantly improves NIK representations. Particularly for high acceleration factors (R$\geq$54), NIK with PISCO achieves superior spatio-temporal reconstruction quality compared to state-of-the-art methods. Furthermore, an extensive analysis of the loss assumptions and stability shows PISCO's potential as versatile self-supervised k-space loss function for further applications and architectures. Code is available at: https://github.com/compai-lab/2025-pisco-spieker
☆ Joint Transmission and Deblurring: A Semantic Communication Approach Using Events
Deep learning-based joint source-channel coding (JSCC) is emerging as a promising technology for effective image transmission. However, most existing approaches focus on transmitting clear images, overlooking real-world challenges such as motion blur caused by camera shaking or fast-moving objects. Motion blur often degrades image quality, making transmission and reconstruction more challenging. Event cameras, which asynchronously record pixel intensity changes with extremely low latency, have shown great potential for motion deblurring tasks. However, the efficient transmission of the abundant data generated by event cameras remains a significant challenge. In this work, we propose a novel JSCC framework for the joint transmission of blurry images and events, aimed at achieving high-quality reconstructions under limited channel bandwidth. This approach is designed as a deblurring task-oriented JSCC system. Since RGB cameras and event cameras capture the same scene through different modalities, their outputs contain both shared and domain-specific information. To avoid repeatedly transmitting the shared information, we extract and transmit their shared information and domain-specific information, respectively. At the receiver, the received signals are processed by a deblurring decoder to generate clear images. Additionally, we introduce a multi-stage training strategy to train the proposed model. Simulation results demonstrate that our method significantly outperforms existing JSCC-based image transmission schemes, addressing motion blur effectively.
☆ SVIA: A Street View Image Anonymization Framework for Self-Driving Applications SC 2024
In recent years, there has been an increasing interest in image anonymization, particularly focusing on the de-identification of faces and individuals. However, for self-driving applications, merely de-identifying faces and individuals might not provide sufficient privacy protection since street views like vehicles and buildings can still disclose locations, trajectories, and other sensitive information. Therefore, it remains crucial to extend anonymization techniques to street view images to fully preserve the privacy of users, pedestrians, and vehicles. In this paper, we propose a Street View Image Anonymization (SVIA) framework for self-driving applications. The SVIA framework consists of three integral components: a semantic segmenter to segment an input image into functional regions, an inpainter to generate alternatives to privacy-sensitive regions, and a harmonizer to seamlessly stitch modified regions to guarantee visual coherence. Compared to existing methods, SVIA achieves a much better trade-off between image generation quality and privacy protection, as evidenced by experimental results for five common metrics on two widely used public datasets.
comment: 8 pages, 6 figures, 3 tables. Accepted by IEEE ITSC 2024
☆ Image Segmentation with transformers: An Overview, Challenges and Future
Image segmentation, a key task in computer vision, has traditionally relied on convolutional neural networks (CNNs), yet these models struggle with capturing complex spatial dependencies, objects with varying scales, need for manually crafted architecture components and contextual information. This paper explores the shortcomings of CNN-based models and the shift towards transformer architectures -to overcome those limitations. This work reviews state-of-the-art transformer-based segmentation models, addressing segmentation-specific challenges and their solutions. The paper discusses current challenges in transformer-based segmentation and outlines promising future trends, such as lightweight architectures and enhanced data efficiency. This survey serves as a guide for understanding the impact of transformers in advancing segmentation capabilities and overcoming the limitations of traditional models.
☆ Identification of Traditional Medicinal Plant Leaves Using an effective Deep Learning model and Self-Curated Dataset
Medicinal plants have been a key component in producing traditional and modern medicines, especially in the field of Ayurveda, an ancient Indian medical system. Producing these medicines and collecting and extracting the right plant is a crucial step due to the visually similar nature of some plants. The extraction of these plants from nonmedicinal plants requires human expert intervention. To solve the issue of accurate plant identification and reduce the need for a human expert in the collection process; employing computer vision methods will be efficient and beneficial. In this paper, we have proposed a model that solves such issues. The proposed model is a custom convolutional neural network (CNN) architecture with 6 convolution layers, max-pooling layers, and dense layers. The model was tested on three different datasets named Indian Medicinal Leaves Image Dataset,MED117 Medicinal Plant Leaf Dataset, and the self-curated dataset by the authors. The proposed model achieved respective accuracies of 99.5%, 98.4%, and 99.7% using various optimizers including Adam, RMSprop, and SGD with momentum.
☆ Strategic Base Representation Learning via Feature Augmentations for Few-Shot Class Incremental Learning WACV 2025
Few-shot class incremental learning implies the model to learn new classes while retaining knowledge of previously learned classes with a small number of training instances. Existing frameworks typically freeze the parameters of the previously learned classes during the incorporation of new classes. However, this approach often results in suboptimal class separation of previously learned classes, leading to overlap between old and new classes. Consequently, the performance of old classes degrades on new classes. To address these challenges, we propose a novel feature augmentation driven contrastive learning framework designed to enhance the separation of previously learned classes to accommodate new classes. Our approach involves augmenting feature vectors and assigning proxy labels to these vectors. This strategy expands the feature space, ensuring seamless integration of new classes within the expanded space. Additionally, we employ a self-supervised contrastive loss to improve the separation between previous classes. We validate our framework through experiments on three FSCIL benchmark datasets: CIFAR100, miniImageNet, and CUB200. The results demonstrate that our Feature Augmentation driven Contrastive Learning framework significantly outperforms other approaches, achieving state-of-the-art performance.
comment: Accepted at WACV 2025
☆ YETI (YET to Intervene) Proactive Interventions by Multimodal AI Agents in Augmented Reality Tasks
Multimodal AI Agents are AI models that have the capability of interactively and cooperatively assisting human users to solve day-to-day tasks. Augmented Reality (AR) head worn devices can uniquely improve the user experience of solving procedural day-to-day tasks by providing egocentric multimodal (audio and video) observational capabilities to AI Agents. Such AR capabilities can help AI Agents see and listen to actions that users take which can relate to multimodal capabilities of human users. Existing AI Agents, either Large Language Models (LLMs) or Multimodal Vision-Language Models (VLMs) are reactive in nature, which means that models cannot take an action without reading or listening to the human user's prompts. Proactivity of AI Agents on the other hand can help the human user detect and correct any mistakes in agent observed tasks, encourage users when they do tasks correctly or simply engage in conversation with the user - akin to a human teaching or assisting a user. Our proposed YET to Intervene (YETI) multimodal agent focuses on the research question of identifying circumstances that may require the agent to intervene proactively. This allows the agent to understand when it can intervene in a conversation with human users that can help the user correct mistakes on tasks, like cooking, using AR. Our YETI Agent learns scene understanding signals based on interpretable notions of Structural Similarity (SSIM) on consecutive video frames. We also define the alignment signal which the AI Agent can learn to identify if the video frames corresponding to the user's actions on the task are consistent with expected actions. These signals are used by our AI Agent to determine when it should proactively intervene. We compare our results on the instances of proactive intervention in the HoloAssist multimodal benchmark for an expert agent guiding a user to complete procedural tasks.
comment: Preprint
☆ Making Your Dreams A Reality: Decoding the Dreams into a Coherent Video Story from fMRI Signals
This paper studies the brave new idea for Multimedia community, and proposes a novel framework to convert dreams into coherent video narratives using fMRI data. Essentially, dreams have intrigued humanity for centuries, offering glimpses into our subconscious minds. Recent advancements in brain imaging, particularly functional magnetic resonance imaging (fMRI), have provided new ways to explore the neural basis of dreaming. By combining subjective dream experiences with objective neurophysiological data, we aim to understand the visual aspects of dreams and create complete video narratives. Our process involves three main steps: reconstructing visual perception, decoding dream imagery, and integrating dream stories. Using innovative techniques in fMRI analysis and language modeling, we seek to push the boundaries of dream research and gain deeper insights into visual experiences during sleep. This technical report introduces a novel approach to visually decoding dreams using fMRI signals and weaving dream visuals into narratives using language models. We gather a dataset of dreams along with descriptions to assess the effectiveness of our framework.
comment: Work in progress
☆ UVRM: A Scalable 3D Reconstruction Model from Unposed Videos
Large Reconstruction Models (LRMs) have recently become a popular method for creating 3D foundational models. Training 3D reconstruction models with 2D visual data traditionally requires prior knowledge of camera poses for the training samples, a process that is both time-consuming and prone to errors. Consequently, 3D reconstruction training has been confined to either synthetic 3D datasets or small-scale datasets with annotated poses. In this study, we investigate the feasibility of 3D reconstruction using unposed video data of various objects. We introduce UVRM, a novel 3D reconstruction model capable of being trained and evaluated on monocular videos without requiring any information about the pose. UVRM uses a transformer network to implicitly aggregate video frames into a pose-invariant latent feature space, which is then decoded into a tri-plane 3D representation. To obviate the need for ground-truth pose annotations during training, UVRM employs a combination of the score distillation sampling (SDS) method and an analysis-by-synthesis approach, progressively synthesizing pseudo novel-views using a pre-trained diffusion model. We qualitatively and quantitatively evaluate UVRM's performance on the G-Objaverse and CO3D datasets without relying on pose information. Extensive experiments show that UVRM is capable of effectively and efficiently reconstructing a wide range of 3D objects from unposed videos.
☆ SE-BSFV: Online Subspace Learning based Shadow Enhancement and Background Suppression for ViSAR under Complex Background
Video synthetic aperture radar (ViSAR) has attracted substantial attention in the moving target detection (MTD) field due to its ability to continuously monitor changes in the target area. In ViSAR, the moving targets' shadows will not offset and defocus, which is widely used as a feature for MTD. However, the shadows are difficult to distinguish from the low scattering region in the background, which will cause more missing and false alarms. Therefore, it is worth investigating how to enhance the distinction between the shadows and background. In this study, we proposed the Shadow Enhancement and Background Suppression for ViSAR (SE-BSFV) algorithm. The SE-BSFV algorithm is based on the low-rank representation (LRR) theory and adopts online subspace learning technique to enhance shadows and suppress background for ViSAR images. Firstly, we use a registration algorithm to register the ViSAR images and utilize Gaussian mixture distribution (GMD) to model the ViSAR data. Secondly, the knowledge learned from the previous frames is leveraged to estimate the GMD parameters of the current frame, and the Expectation-maximization (EM) algorithm is used to estimate the subspace parameters. Then, the foreground matrix of the current frame can be obtained. Finally, the alternating direction method of multipliers (ADMM) is used to eliminate strong scattering objects in the foreground matrix to obtain the final results. The experimental results indicate that the SE-BSFV algorithm significantly enhances the shadows' saliency and greatly improves the detection performance while ensuring efficiency compared with several other advanced pre-processing algorithms.
Prompt-CAM: A Simpler Interpretable Transformer for Fine-Grained Analysis
We present a simple usage of pre-trained Vision Transformers (ViTs) for fine-grained analysis, aiming to identify and localize the traits that distinguish visually similar categories, such as different bird species or dog breeds. Pre-trained ViTs such as DINO have shown remarkable capabilities to extract localized, informative features. However, using saliency maps like Grad-CAM can hardly point out the traits: they often locate the whole object by a blurred, coarse heatmap, not traits. We propose a novel approach Prompt Class Attention Map (Prompt-CAM) to the rescue. Prompt-CAM learns class-specific prompts to a pre-trained ViT and uses the corresponding outputs for classification. To classify an image correctly, the true-class prompt must attend to the unique image patches not seen in other classes' images, i.e., traits. As such, the true class's multi-head attention maps reveal traits and their locations. Implementation-wise, Prompt-CAM is almost a free lunch by simply modifying the prediction head of Visual Prompt Tuning (VPT). This makes Prompt-CAM fairly easy to train and apply, sharply contrasting other interpretable methods that design specific models and training processes. It is even simpler than the recently published INterpretable TRansformer (INTR), whose encoder-decoder architecture prevents it from leveraging pre-trained ViTs. Extensive empirical studies on a dozen datasets from various domains (e.g., birds, fishes, insects, fungi, flowers, food, and cars) validate Prompt-CAM superior interpretation capability.
☆ Soft Knowledge Distillation with Multi-Dimensional Cross-Net Attention for Image Restoration Models Compression ICASSP2025
Transformer-based encoder-decoder models have achieved remarkable success in image-to-image transfer tasks, particularly in image restoration. However, their high computational complexity-manifested in elevated FLOPs and parameter counts-limits their application in real-world scenarios. Existing knowledge distillation methods in image restoration typically employ lightweight student models that directly mimic the intermediate features and reconstruction results of the teacher, overlooking the implicit attention relationships between them. To address this, we propose a Soft Knowledge Distillation (SKD) strategy that incorporates a Multi-dimensional Cross-net Attention (MCA) mechanism for compressing image restoration models. This mechanism facilitates interaction between the student and teacher across both channel and spatial dimensions, enabling the student to implicitly learn the attention matrices. Additionally, we employ a Gaussian kernel function to measure the distance between student and teacher features in kernel space, ensuring stable and efficient feature learning. To further enhance the quality of reconstructed images, we replace the commonly used L1 or KL divergence loss with a contrastive learning loss at the image level. Experiments on three tasks-image deraining, deblurring, and denoising-demonstrate that our SKD strategy significantly reduces computational complexity while maintaining strong image restoration capabilities.
comment: Accepted by ICASSP2025
☆ Shape-Based Single Object Classification Using Ensemble Method Classifiers
Nowadays, more and more images are available. Annotation and retrieval of the images pose classification problems, where each class is defined as the group of database images labelled with a common semantic label. Various systems have been proposed for content-based retrieval, as well as for image classification and indexing. In this paper, a hierarchical classification framework has been proposed for bridging the semantic gap effectively and achieving multi-category image classification. A well known pre-processing and post-processing method was used and applied to three problems; image segmentation, object identification and image classification. The method was applied to classify single object images from Amazon and Google datasets. The classification was tested for four different classifiers; BayesNetwork (BN), Random Forest (RF), Bagging and Vote. The estimated classification accuracies ranged from 20% to 99% (using 10-fold cross validation). The Bagging classifier presents the best performance, followed by the Random Forest classifier.
☆ Domain-conditioned and Temporal-guided Diffusion Modeling for Accelerated Dynamic MRI Reconstruction
Purpose: To propose a domain-conditioned and temporal-guided diffusion modeling method, termed dynamic Diffusion Modeling (dDiMo), for accelerated dynamic MRI reconstruction, enabling diffusion process to characterize spatiotemporal information for time-resolved multi-coil Cartesian and non-Cartesian data. Methods: The dDiMo framework integrates temporal information from time-resolved dimensions, allowing for the concurrent capture of intra-frame spatial features and inter-frame temporal dynamics in diffusion modeling. It employs additional spatiotemporal ($x$-$t$) and self-consistent frequency-temporal ($k$-$t$) priors to guide the diffusion process. This approach ensures precise temporal alignment and enhances the recovery of fine image details. To facilitate a smooth diffusion process, the nonlinear conjugate gradient algorithm is utilized during the reverse diffusion steps. The proposed model was tested on two types of MRI data: Cartesian-acquired multi-coil cardiac MRI and Golden-Angle-Radial-acquired multi-coil free-breathing lung MRI, across various undersampling rates. Results: dDiMo achieved high-quality reconstructions at various acceleration factors, demonstrating improved temporal alignment and structural recovery compared to other competitive reconstruction methods, both qualitatively and quantitatively. This proposed diffusion framework exhibited robust performance in handling both Cartesian and non-Cartesian acquisitions, effectively reconstructing dynamic datasets in cardiac and lung MRI under different imaging conditions. Conclusion: This study introduces a novel diffusion modeling method for dynamic MRI reconstruction.
comment: 21 pages, 15 figures, 2 tables
☆ Finding the Trigger: Causal Abductive Reasoning on Video Events
This paper introduces a new problem, Causal Abductive Reasoning on Video Events (CARVE), which involves identifying causal relationships between events in a video and generating hypotheses about causal chains that account for the occurrence of a target event. To facilitate research in this direction, we create two new benchmark datasets with both synthetic and realistic videos, accompanied by trigger-target labels generated through a novel counterfactual synthesis approach. To explore the challenge of solving CARVE, we present a Causal Event Relation Network (CERN) that examines the relationships between video events in temporal and semantic spaces to efficiently determine the root-cause trigger events. Through extensive experiments, we demonstrate the critical roles of event relational representation learning and interaction modeling in solving video causal reasoning challenges. The introduction of the CARVE task, along with the accompanying datasets and the CERN framework, will advance future research on video causal reasoning and significantly facilitate various applications, including video surveillance, root-cause analysis and movie content management.
☆ Creating Virtual Environments with 3D Gaussian Splatting: A Comparative Study
3D Gaussian Splatting (3DGS) has recently emerged as an innovative and efficient 3D representation technique. While its potential for extended reality (XR) applications is frequently highlighted, its practical effectiveness remains underexplored. In this work, we examine three distinct 3DGS-based approaches for virtual environment (VE) creation, leveraging their unique strengths for efficient and visually compelling scene representation. By conducting a comparable study, we evaluate the feasibility of 3DGS in creating immersive VEs, identify its limitations in XR applications, and discuss future research and development opportunities.
comment: IEEE VR 2025 Posters
☆ Efficient Few-Shot Medical Image Analysis via Hierarchical Contrastive Vision-Language Learning
Few-shot learning in medical image classification presents a significant challenge due to the limited availability of annotated data and the complex nature of medical imagery. In this work, we propose Adaptive Vision-Language Fine-tuning with Hierarchical Contrastive Alignment (HiCA), a novel framework that leverages the capabilities of Large Vision-Language Models (LVLMs) for medical image analysis. HiCA introduces a two-stage fine-tuning strategy, combining domain-specific pretraining and hierarchical contrastive learning to align visual and textual representations at multiple levels. We evaluate our approach on two benchmark datasets, Chest X-ray and Breast Ultrasound, achieving state-of-the-art performance in both few-shot and zero-shot settings. Further analyses demonstrate the robustness, generalizability, and interpretability of our method, with substantial improvements in performance compared to existing baselines. Our work highlights the potential of hierarchical contrastive strategies in adapting LVLMs to the unique challenges of medical imaging tasks.
☆ SoccerSynth-Detection: A Synthetic Dataset for Soccer Player Detection
In soccer video analysis, player detection is essential for identifying key events and reconstructing tactical positions. The presence of numerous players and frequent occlusions, combined with copyright restrictions, severely restricts the availability of datasets, leaving limited options such as SoccerNet-Tracking and SportsMOT. These datasets suffer from a lack of diversity, which hinders algorithms from adapting effectively to varied soccer video contexts. To address these challenges, we developed SoccerSynth-Detection, the first synthetic dataset designed for the detection of synthetic soccer players. It includes a broad range of random lighting and textures, as well as simulated camera motion blur. We validated its efficacy using the object detection model (Yolov8n) against real-world datasets (SoccerNet-Tracking and SportsMoT). In transfer tests, it matched the performance of real datasets and significantly outperformed them in images with motion blur; in pre-training tests, it demonstrated its efficacy as a pre-training dataset, significantly enhancing the algorithm's overall performance. Our work demonstrates the potential of synthetic datasets to replace real datasets for algorithm training in the field of soccer video analysis.
☆ Text-guided Synthetic Geometric Augmentation for Zero-shot 3D Understanding CVPR
Zero-shot recognition models require extensive training data for generalization. However, in zero-shot 3D classification, collecting 3D data and captions is costly and laborintensive, posing a significant barrier compared to 2D vision. Recent advances in generative models have achieved unprecedented realism in synthetic data production, and recent research shows the potential for using generated data as training data. Here, naturally raising the question: Can synthetic 3D data generated by generative models be used as expanding limited 3D datasets? In response, we present a synthetic 3D dataset expansion method, Textguided Geometric Augmentation (TeGA). TeGA is tailored for language-image-3D pretraining, which achieves SoTA in zero-shot 3D classification, and uses a generative textto-3D model to enhance and extend limited 3D datasets. Specifically, we automatically generate text-guided synthetic 3D data and introduce a consistency filtering strategy to discard noisy samples where semantics and geometric shapes do not match with text. In the experiment to double the original dataset size using TeGA, our approach demonstrates improvements over the baselines, achieving zeroshot performance gains of 3.0% on Objaverse-LVIS, 4.6% on ScanObjectNN, and 8.7% on ModelNet40. These results demonstrate that TeGA effectively bridges the 3D data gap, enabling robust zero-shot 3D classification even with limited real training data and paving the way for zero-shot 3D vision application.
comment: 14 pages, 8 figures, this paper is submitted to CVPR
☆ Bias for Action: Video Implicit Neural Representations with Bias Modulation
We propose a new continuous video modeling framework based on implicit neural representations (INRs) called ActINR. At the core of our approach is the observation that INRs can be considered as a learnable dictionary, with the shapes of the basis functions governed by the weights of the INR, and their locations governed by the biases. Given compact non-linear activation functions, we hypothesize that an INR's biases are suitable to capture motion across images, and facilitate compact representations for video sequences. Using these observations, we design ActINR to share INR weights across frames of a video sequence, while using unique biases for each frame. We further model the biases as the output of a separate INR conditioned on time index to promote smoothness. By training the video INR and this bias INR together, we demonstrate unique capabilities, including $10\times$ video slow motion, $4\times$ spatial super resolution along with $2\times$ slow motion, denoising, and video inpainting. ActINR performs remarkably well across numerous video processing tasks (often achieving more than 6dB improvement), setting a new standard for continuous modeling of videos.
☆ Knowledge Distillation for Image Restoration : Simultaneous Learning from Degraded and Clean Images ICASSP2025
Model compression through knowledge distillation has seen extensive application in classification and segmentation tasks. However, its potential in image-to-image translation, particularly in image restoration, remains underexplored. To address this gap, we propose a Simultaneous Learning Knowledge Distillation (SLKD) framework tailored for model compression in image restoration tasks. SLKD employs a dual-teacher, single-student architecture with two distinct learning strategies: Degradation Removal Learning (DRL) and Image Reconstruction Learning (IRL), simultaneously. In DRL, the student encoder learns from Teacher A to focus on removing degradation factors, guided by a novel BRISQUE extractor. In IRL, the student decoder learns from Teacher B to reconstruct clean images, with the assistance of a proposed PIQE extractor. These strategies enable the student to learn from degraded and clean images simultaneously, ensuring high-quality compression of image restoration models. Experimental results across five datasets and three tasks demonstrate that SLKD achieves substantial reductions in FLOPs and parameters, exceeding 80\%, while maintaining strong image restoration performance.
comment: Accepted by ICASSP2025
☆ Are Open-Vocabulary Models Ready for Detection of MEP Elements on Construction Sites
The construction industry has long explored robotics and computer vision, yet their deployment on construction sites remains very limited. These technologies have the potential to revolutionize traditional workflows by enhancing accuracy, efficiency, and safety in construction management. Ground robots equipped with advanced vision systems could automate tasks such as monitoring mechanical, electrical, and plumbing (MEP) systems. The present research evaluates the applicability of open-vocabulary vision-language models compared to fine-tuned, lightweight, closed-set object detectors for detecting MEP components using a mobile ground robotic platform. A dataset collected with cameras mounted on a ground robot was manually annotated and analyzed to compare model performance. The results demonstrate that, despite the versatility of vision-language models, fine-tuned lightweight models still largely outperform them in specialized environments and for domain-specific tasks.
comment: 4 pages, 3 figures
☆ OpticFusion: Multi-Modal Neural Implicit 3D Reconstruction of Microstructures by Fusing White Light Interferometry and Optical Microscopy 3DV 2025
White Light Interferometry (WLI) is a precise optical tool for measuring the 3D topography of microstructures. However, conventional WLI cannot capture the natural color of a sample's surface, which is essential for many microscale research applications that require both 3D geometry and color information. Previous methods have attempted to overcome this limitation by modifying WLI hardware and analysis software, but these solutions are often costly. In this work, we address this challenge from a computer vision multi-modal reconstruction perspective for the first time. We introduce OpticFusion, a novel approach that uses an additional digital optical microscope (OM) to achieve 3D reconstruction with natural color textures using multi-view WLI and OM images. Our method employs a two-step data association process to obtain the poses of WLI and OM data. By leveraging the neural implicit representation, we fuse multi-modal data and apply color decomposition technology to extract the sample's natural color. Tested on our multi-modal dataset of various microscale samples, OpticFusion achieves detailed 3D reconstructions with color textures. Our method provides an effective tool for practical applications across numerous microscale research fields. The source code and our real-world dataset are available at https://github.com/zju3dv/OpticFusion.
comment: 3DV 2025
☆ Leveraging Scale-aware Representations for improved Concept-Representation Alignment in ViTs
Vision Transformers (ViTs) are increasingly being adopted in various sensitive vision applications - like medical diagnosis, facial recognition, etc. To improve the interpretability of such models, many approaches attempt to forward-align them with carefully annotated abstract, human-understandable semantic entities - concepts. Concepts provide global rationales to the model predictions and can be quickly understood/intervened on by domain experts. Most current research focuses on designing model-agnostic, plug-and-play generic concept-based explainability modules that do not incorporate the inner workings of foundation models (e.g., inductive biases, scale invariance, etc.) during training. To alleviate this issue for ViTs, in this paper, we propose a novel Concept Representation Alignment Module (CRAM) which learns both scale and position-aware representations from multi-scale feature pyramids and patch representations respectively. CRAM further aligns these representations with concept annotations through an attention matrix. The proposed CRAM module improves the predictive performance of ViT architectures and also provides accurate and robust concept explanations as demonstrated on five datasets - including three widely used benchmarks (CUB, Pascal APY, Concept-MNIST) and 2 real-world datasets (AWA2, KITS).
☆ Adaptive Law-Based Transformation (ALT): A Lightweight Feature Representation for Time Series Classification
Time series classification (TSC) is fundamental in numerous domains, including finance, healthcare, and environmental monitoring. However, traditional TSC methods often struggle with the inherent complexity and variability of time series data. Building on our previous work with the linear law-based transformation (LLT) - which improved classification accuracy by transforming the feature space based on key data patterns - we introduce adaptive law-based transformation (ALT). ALT enhances LLT by incorporating variable-length shifted time windows, enabling it to capture distinguishing patterns of various lengths and thereby handle complex time series more effectively. By mapping features into a linearly separable space, ALT provides a fast, robust, and transparent solution that achieves state-of-the-art performance with only a few hyperparameters.
comment: 8 pages, 1 figure, 5 tables
☆ Surgical Visual Understanding (SurgVU) Dataset
Owing to recent advances in machine learning and the ability to harvest large amounts of data during robotic-assisted surgeries, surgical data science is ripe for foundational work. We present a large dataset of surgical videos and their accompanying labels for this purpose. We describe how the data was collected and some of its unique attributes. Multiple example problems are outlined. Although the dataset was curated for a particular set of scientific challenges (in an accompanying paper), it is general enough to be used for a broad range machine learning questions. Our hope is that this dataset exposes the larger machine learning community to the challenging problems within surgical data science, and becomes a touchstone for future research. The videos are available at https://storage.googleapis.com/isi-surgvu/surgvu24_videos_only.zip, the labels at https://storage.googleapis.com/isi-surgvu/surgvu24_labels_updated_v2.zip, and a validation set for tool detection problem at https://storage.googleapis.com/isi-surgvu/cat1_test_set_public.zip.
☆ Semi-Supervised Image-Based Narrative Extraction: A Case Study with Historical Photographic Records ECIR 2025
This paper presents a semi-supervised approach to extracting narratives from historical photographic records using an adaptation of the narrative maps algorithm. We extend the original unsupervised text-based method to work with image data, leveraging deep learning techniques for visual feature extraction and similarity computation. Our method is applied to the ROGER dataset, a collection of photographs from the 1928 Sacambaya Expedition in Bolivia captured by Robert Gerstmann. We compare our algorithmically extracted visual narratives with expert-curated timelines of varying lengths (5 to 30 images) to evaluate the effectiveness of our approach. In particular, we use the Dynamic Time Warping (DTW) algorithm to match the extracted narratives with the expert-curated baseline. In addition, we asked an expert on the topic to qualitatively evaluate a representative example of the resulting narratives. Our findings show that the narrative maps approach generally outperforms random sampling for longer timelines (10+ images, p < 0.05), with expert evaluation confirming the historical accuracy and coherence of the extracted narratives. This research contributes to the field of computational analysis of visual cultural heritage, offering new tools for historians, archivists, and digital humanities scholars to explore and understand large-scale image collections. The method's ability to generate meaningful narratives from visual data opens up new possibilities for the study and interpretation of historical events through photographic evidence.
comment: This paper has been accepted for oral presentation in the findings track of the 47th European Conference on Information Retrieval (ECIR 2025). Source code and experiments are available at https://github.com/faustogerman/ROGER-Concept-Narratives
☆ ASTRA: A Scene-aware TRAnsformer-based model for trajectory prediction
We present ASTRA (A} Scene-aware TRAnsformer-based model for trajectory prediction), a light-weight pedestrian trajectory forecasting model that integrates the scene context, spatial dynamics, social inter-agent interactions and temporal progressions for precise forecasting. We utilised a U-Net-based feature extractor, via its latent vector representation, to capture scene representations and a graph-aware transformer encoder for capturing social interactions. These components are integrated to learn an agent-scene aware embedding, enabling the model to learn spatial dynamics and forecast the future trajectory of pedestrians. The model is designed to produce both deterministic and stochastic outcomes, with the stochastic predictions being generated by incorporating a Conditional Variational Auto-Encoder (CVAE). ASTRA also proposes a simple yet effective weighted penalty loss function, which helps to yield predictions that outperform a wide array of state-of-the-art deterministic and generative models. ASTRA demonstrates an average improvement of 27%/10% in deterministic/stochastic settings on the ETH-UCY dataset, and 26% improvement on the PIE dataset, respectively, along with seven times fewer parameters than the existing state-of-the-art model (see Figure 1). Additionally, the model's versatility allows it to generalize across different perspectives, such as Bird's Eye View (BEV) and Ego-Vehicle View (EVV).
☆ Detection of Vascular Leukoencephalopathy in CT Images
Artificial intelligence (AI) has seen a significant surge in popularity, particularly in its application to medicine. This study explores AI's role in diagnosing leukoencephalopathy, a small vessel disease of the brain, and a leading cause of vascular dementia and hemorrhagic strokes. We utilized a dataset of approximately 1200 patients with axial brain CT scans to train convolutional neural networks (CNNs) for binary disease classification. Addressing the challenge of varying scan dimensions due to different patient physiologies, we processed the data to a uniform size and applied three preprocessing methods to improve model accuracy. We compared four neural network architectures: ResNet50, ResNet50 3D, ConvNext, and Densenet. The ConvNext model achieved the highest accuracy of 98.5% without any preprocessing, outperforming models with 3D convolutions. To gain insights into model decision-making, we implemented Grad-CAM heatmaps, which highlighted the focus areas of the models on the scans. Our results demonstrate that AI, particularly the ConvNext architecture, can significantly enhance diagnostic accuracy for leukoencephalopathy. This study underscores AI's potential in advancing diagnostic methodologies for brain diseases and highlights the effectiveness of CNNs in medical imaging applications.
☆ CrossModalityDiffusion: Multi-Modal Novel View Synthesis with Unified Intermediate Representation WACV
Geospatial imaging leverages data from diverse sensing modalities-such as EO, SAR, and LiDAR, ranging from ground-level drones to satellite views. These heterogeneous inputs offer significant opportunities for scene understanding but present challenges in interpreting geometry accurately, particularly in the absence of precise ground truth data. To address this, we propose CrossModalityDiffusion, a modular framework designed to generate images across different modalities and viewpoints without prior knowledge of scene geometry. CrossModalityDiffusion employs modality-specific encoders that take multiple input images and produce geometry-aware feature volumes that encode scene structure relative to their input camera positions. The space where the feature volumes are placed acts as a common ground for unifying input modalities. These feature volumes are overlapped and rendered into feature images from novel perspectives using volumetric rendering techniques. The rendered feature images are used as conditioning inputs for a modality-specific diffusion model, enabling the synthesis of novel images for the desired output modality. In this paper, we show that jointly training different modules ensures consistent geometric understanding across all modalities within the framework. We validate CrossModalityDiffusion's capabilities on the synthetic ShapeNet cars dataset, demonstrating its effectiveness in generating accurate and consistent novel views across multiple imaging modalities and perspectives.
comment: Accepted in the 2025 WACV workshop GeoCV
☆ EraseBench: Understanding The Ripple Effects of Concept Erasure Techniques
Concept erasure techniques have recently gained significant attention for their potential to remove unwanted concepts from text-to-image models. While these methods often demonstrate success in controlled scenarios, their robustness in real-world applications and readiness for deployment remain uncertain. In this work, we identify a critical gap in evaluating sanitized models, particularly in terms of their performance across various concept dimensions. We systematically investigate the failure modes of current concept erasure techniques, with a focus on visually similar, binomial, and semantically related concepts. We propose that these interconnected relationships give rise to a phenomenon of concept entanglement resulting in ripple effects and degradation in image quality. To facilitate more comprehensive evaluation, we introduce EraseBENCH, a multi-dimensional benchmark designed to assess concept erasure methods with greater depth. Our dataset includes over 100 diverse concepts and more than 1,000 tailored prompts, paired with a comprehensive suite of metrics that together offer a holistic view of erasure efficacy. Our findings reveal that even state-of-the-art techniques struggle with maintaining quality post-erasure, indicating that these approaches are not yet ready for real-world deployment. This highlights the gap in reliability of the concept erasure techniques.
comment: 11 pages main; 9 pages supplemental material
☆ PIXELS: Progressive Image Xemplar-based Editing with Latent Surgery
Recent advancements in language-guided diffusion models for image editing are often bottle-necked by cumbersome prompt engineering to precisely articulate desired changes. An intuitive alternative calls on guidance from in-the-wild image exemplars to help users bring their imagined edits to life. Contemporary exemplar-based editing methods shy away from leveraging the rich latent space learnt by pre-existing large text-to-image (TTI) models and fall back on training with curated objective functions to achieve the task. Though somewhat effective, this demands significant computational resources and lacks compatibility with diverse base models and arbitrary exemplar count. On further investigation, we also find that these techniques restrict user control to only applying uniform global changes over the entire edited region. In this paper, we introduce a novel framework for progressive exemplar-driven editing with off-the-shelf diffusion models, dubbed PIXELS, to enable customization by providing granular control over edits, allowing adjustments at the pixel or region level. Our method operates solely during inference to facilitate imitative editing, enabling users to draw inspiration from a dynamic number of reference images, or multimodal prompts, and progressively incorporate all the desired changes without retraining or fine-tuning existing TTI models. This capability of fine-grained control opens up a range of new possibilities, including selective modification of individual objects and specifying gradual spatial changes. We demonstrate that PIXELS delivers high-quality edits efficiently, leading to a notable improvement in quantitative metrics as well as human evaluation. By making high-quality image editing more accessible, PIXELS has the potential to enable professional-grade edits to a wider audience with the ease of using any open-source image generation model.
☆ Generalized Single-Image-Based Morphing Attack Detection Using Deep Representations from Vision Transformer
Face morphing attacks have posed severe threats to Face Recognition Systems (FRS), which are operated in border control and passport issuance use cases. Correspondingly, morphing attack detection algorithms (MAD) are needed to defend against such attacks. MAD approaches must be robust enough to handle unknown attacks in an open-set scenario where attacks can originate from various morphing generation algorithms, post-processing and the diversity of printers/scanners. The problem of generalization is further pronounced when the detection has to be made on a single suspected image. In this paper, we propose a generalized single-image-based MAD (S-MAD) algorithm by learning the encoding from Vision Transformer (ViT) architecture. Compared to CNN-based architectures, ViT model has the advantage on integrating local and global information and hence can be suitable to detect the morphing traces widely distributed among the face region. Extensive experiments are carried out on face morphing datasets generated using publicly available FRGC face datasets. Several state-of-the-art (SOTA) MAD algorithms, including representative ones that have been publicly evaluated, have been selected and benchmarked with our ViT-based approach. Obtained results demonstrate the improved detection performance of the proposed S-MAD method on inter-dataset testing (when different data is used for training and testing) and comparable performance on intra-dataset testing (when the same data is used for training and testing) experimental protocol.
☆ Lossy Compression with Pretrained Diffusion Models
We apply the DiffC algorithm (Theis et al. 2022) to Stable Diffusion 1.5, 2.1, XL, and Flux-dev, and demonstrate that these pretrained models are remarkably capable lossy image compressors. A principled algorithm for lossy compression using pretrained diffusion models has been understood since at least Ho et al. 2020, but challenges in reverse-channel coding have prevented such algorithms from ever being fully implemented. We introduce simple workarounds that lead to the first complete implementation of DiffC, which is capable of compressing and decompressing images using Stable Diffusion in under 10 seconds. Despite requiring no additional training, our method is competitive with other state-of-the-art generative compression methods at low ultra-low bitrates.
☆ SMPLest-X: Ultimate Scaling for Expressive Human Pose and Shape Estimation
Expressive human pose and shape estimation (EHPS) unifies body, hands, and face motion capture with numerous applications. Despite encouraging progress, current state-of-the-art methods focus on training innovative architectural designs on confined datasets. In this work, we investigate the impact of scaling up EHPS towards a family of generalist foundation models. 1) For data scaling, we perform a systematic investigation on 40 EHPS datasets, encompassing a wide range of scenarios that a model trained on any single dataset cannot handle. More importantly, capitalizing on insights obtained from the extensive benchmarking process, we optimize our training scheme and select datasets that lead to a significant leap in EHPS capabilities. Ultimately, we achieve diminishing returns at 10M training instances from diverse data sources. 2) For model scaling, we take advantage of vision transformers (up to ViT-Huge as the backbone) to study the scaling law of model sizes in EHPS. To exclude the influence of algorithmic design, we base our experiments on two minimalist architectures: SMPLer-X, which consists of an intermediate step for hand and face localization, and SMPLest-X, an even simpler version that reduces the network to its bare essentials and highlights significant advances in the capture of articulated hands. With big data and the large model, the foundation models exhibit strong performance across diverse test benchmarks and excellent transferability to even unseen environments. Moreover, our finetuning strategy turns the generalist into specialist models, allowing them to achieve further performance boosts. Notably, our foundation models consistently deliver state-of-the-art results on seven benchmarks such as AGORA, UBody, EgoBody, and our proposed SynHand dataset for comprehensive hand evaluation. (Code is available at: https://github.com/wqyin/SMPLest-X).
comment: An extension of SMPLer-X [arXiv:2309.17448]. Homepage: https://caizhongang.com/projects/SMPLer-X/
☆ VideoWorld: Exploring Knowledge Learning from Unlabeled Videos
This work explores whether a deep generative model can learn complex knowledge solely from visual input, in contrast to the prevalent focus on text-based models like large language models (LLMs). We develop VideoWorld, an auto-regressive video generation model trained on unlabeled video data, and test its knowledge acquisition abilities in video-based Go and robotic control tasks. Our experiments reveal two key findings: (1) video-only training provides sufficient information for learning knowledge, including rules, reasoning and planning capabilities, and (2) the representation of visual change is crucial for knowledge acquisition. To improve both the efficiency and efficacy of this process, we introduce the Latent Dynamics Model (LDM) as a key component of VideoWorld. Remarkably, VideoWorld reaches a 5-dan professional level in the Video-GoBench with just a 300-million-parameter model, without relying on search algorithms or reward mechanisms typical in reinforcement learning. In robotic tasks, VideoWorld effectively learns diverse control operations and generalizes across environments, approaching the performance of oracle models in CALVIN and RLBench. This study opens new avenues for knowledge acquisition from visual data, with all code, data, and models open-sourced for further research.
comment: Code and models are released at: https://maverickren.github.io/VideoWorld.github.io/
♻ ☆ FutureDepth: Learning to Predict the Future Improves Video Depth Estimation ECCV 2024
In this paper, we propose a novel video depth estimation approach, FutureDepth, which enables the model to implicitly leverage multi-frame and motion cues to improve depth estimation by making it learn to predict the future at training. More specifically, we propose a future prediction network, F-Net, which takes the features of multiple consecutive frames and is trained to predict multi-frame features one time step ahead iteratively. In this way, F-Net learns the underlying motion and correspondence information, and we incorporate its features into the depth decoding process. Additionally, to enrich the learning of multiframe correspondence cues, we further leverage a reconstruction network, R-Net, which is trained via adaptively masked auto-encoding of multiframe feature volumes. At inference time, both F-Net and R-Net are used to produce queries to work with the depth decoder, as well as a final refinement network. Through extensive experiments on several benchmarks, i.e., NYUDv2, KITTI, DDAD, and Sintel, which cover indoor, driving, and open-domain scenarios, we show that FutureDepth significantly improves upon baseline models, outperforms existing video depth estimation methods, and sets new state-of-the-art (SOTA) accuracy. Furthermore, FutureDepth is more efficient than existing SOTA video depth estimation models and has similar latencies when comparing to monocular models
comment: ECCV 2024
♻ ☆ MAMo: Leveraging Memory and Attention for Monocular Video Depth Estimation ICCV 2023
We propose MAMo, a novel memory and attention frame-work for monocular video depth estimation. MAMo can augment and improve any single-image depth estimation networks into video depth estimation models, enabling them to take advantage of the temporal information to predict more accurate depth. In MAMo, we augment model with memory which aids the depth prediction as the model streams through the video. Specifically, the memory stores learned visual and displacement tokens of the previous time instances. This allows the depth network to cross-reference relevant features from the past when predicting depth on the current frame. We introduce a novel scheme to continuously update the memory, optimizing it to keep tokens that correspond with both the past and the present visual information. We adopt attention-based approach to process memory features where we first learn the spatio-temporal relation among the resultant visual and displacement memory tokens using self-attention module. Further, the output features of self-attention are aggregated with the current visual features through cross-attention. The cross-attended features are finally given to a decoder to predict depth on the current frame. Through extensive experiments on several benchmarks, including KITTI, NYU-Depth V2, and DDAD, we show that MAMo consistently improves monocular depth estimation networks and sets new state-of-the-art (SOTA) accuracy. Notably, our MAMo video depth estimation provides higher accuracy with lower latency, when omparing to SOTA cost-volume-based video depth models.
comment: Accepted at ICCV 2023
♻ ☆ Vulnerability-Aware Spatio-Temporal Learning for Generalizable and Interpretable Deepfake Video Detection
Detecting deepfake videos is highly challenging due to the complex intertwined spatial and temporal artifacts in forged sequences. Most recent approaches rely on binary classifiers trained on both real and fake data. However, such methods may struggle to focus on important artifacts, which can hinder their generalization capability. Additionally, these models often lack interpretability, making it difficult to understand how predictions are made. To address these issues, we propose FakeSTormer, offering two key contributions. First, we introduce a multi-task learning framework with additional spatial and temporal branches that enable the model to focus on subtle spatio-temporal artifacts. These branches also provide interpretability by highlighting video regions that may contain artifacts. Second, we propose a video-level data synthesis algorithm that generates pseudo-fake videos with subtle artifacts, providing the model with high-quality samples and ground truth data for our spatial and temporal branches. Extensive experiments on several challenging benchmarks demonstrate the competitiveness of our approach compared to recent state-of-the-art methods. The code is available at https://github.com/10Ring/FakeSTormer.
♻ ☆ Super-class guided Transformer for Zero-Shot Attribute Classification AAAI25
Attribute classification is crucial for identifying specific characteristics within image regions. Vision-Language Models (VLMs) have been effective in zero-shot tasks by leveraging their general knowledge from large-scale datasets. Recent studies demonstrate that transformer-based models with class-wise queries can effectively address zero-shot multi-label classification. However, poor utilization of the relationship between seen and unseen attributes makes the model lack generalizability. Additionally, attribute classification generally involves many attributes, making maintaining the model's scalability difficult. To address these issues, we propose Super-class guided transFormer (SugaFormer), a novel framework that leverages super-classes to enhance scalability and generalizability for zero-shot attribute classification. SugaFormer employs Super-class Query Initialization (SQI) to reduce the number of queries, utilizing common semantic information from super-classes, and incorporates Multi-context Decoding (MD) to handle diverse visual cues. To strengthen generalizability, we introduce two knowledge transfer strategies that utilize VLMs. During training, Super-class guided Consistency Regularization (SCR) aligns model's features with VLMs using super-class guided prompts, and during inference, Zero-shot Retrieval-based Score Enhancement (ZRSE) refines predictions for unseen attributes. Extensive experiments demonstrate that SugaFormer achieves state-of-the-art performance across three widely-used attribute classification benchmarks under zero-shot, and cross-dataset transfer settings. Our code is available at https://github.com/mlvlab/SugaFormer.
comment: AAAI25
♻ ☆ VIS-MAE: An Efficient Self-supervised Learning Approach on Medical Image Segmentation and Classification
Artificial Intelligence (AI) has the potential to revolutionize diagnosis and segmentation in medical imaging. However, development and clinical implementation face multiple challenges including limited data availability, lack of generalizability, and the necessity to incorporate multi-modal data effectively. A foundation model, which is a large-scale pre-trained AI model, offers a versatile base that can be adapted to a variety of specific tasks and contexts. Here, we present VIsualization and Segmentation Masked AutoEncoder (VIS-MAE), novel model weights specifically designed for medical imaging. Specifically, VIS-MAE is trained on a dataset of 2.5 million unlabeled images from various modalities (CT, MR, PET,X-rays, and ultrasound), using self-supervised learning techniques. It is then adapted to classification and segmentation tasks using explicit labels. VIS-MAE has high label efficiency, outperforming several benchmark models in both in-domain and out-of-domain applications. In addition, VIS-MAE has improved label efficiency as it can achieve similar performance to other models with a reduced amount of labeled training data (50% or 80%) compared to other pre-trained weights. VIS-MAE represents a significant advancement in medical imaging AI, offering a generalizable and robust solution for improving segmentation and classification tasks while reducing the data annotation workload. The source code of this work is available at https://github.com/lzl199704/VIS-MAE.
♻ ☆ A Comparative Study on Multi-task Uncertainty Quantification in Semantic Segmentation and Monocular Depth Estimation
Deep neural networks excel in perception tasks such as semantic segmentation and monocular depth estimation, making them indispensable in safety-critical applications like autonomous driving and industrial inspection. However, they often suffer from overconfidence and poor explainability, especially for out-of-domain data. While uncertainty quantification has emerged as a promising solution to these challenges, multi-task settings have yet to be explored. In an effort to shed light on this, we evaluate Monte Carlo Dropout, Deep Sub-Ensembles, and Deep Ensembles for joint semantic segmentation and monocular depth estimation. Thereby, we reveal that Deep Ensembles stand out as the preferred choice, particularly in out-of-domain scenarios, and show the potential benefit of multi-task learning with regard to the uncertainty quality in comparison to solving both tasks separately. Additionally, we highlight the impact of employing different uncertainty thresholds to classify pixels as certain or uncertain, with the median uncertainty emerging as a robust default.
comment: This manuscript is an extended version of a previously published conference paper and is currently in review for a journal
♻ ☆ A Comprehensive Survey of Foundation Models in Medicine
Foundation models (FMs) are large-scale deep learning models trained on massive datasets, often using self-supervised learning techniques. These models serve as a versatile base for a wide range of downstream tasks, including those in medicine and healthcare. FMs have demonstrated remarkable success across multiple healthcare domains. However, existing surveys in this field do not comprehensively cover all areas where FMs have made significant strides. In this survey, we present a comprehensive review of FMs in medicine, focusing on their evolution, learning strategies, flagship models, applications, and associated challenges. We examine how prominent FMs, such as the BERT and GPT families, are transforming various aspects of healthcare, including clinical large language models, medical image analysis, and omics research. Additionally, we provide a detailed taxonomy of FM-enabled healthcare applications, spanning clinical natural language processing, medical computer vision, graph learning, and other biology- and omics- related tasks. Despite the transformative potentials of FMs, they also pose unique challenges. This survey delves into these challenges and highlights open research questions and lessons learned to guide researchers and practitioners. Our goal is to provide valuable insights into the capabilities of FMs in health, facilitating responsible deployment and mitigating associated risks.
comment: Currently under review in IEEE REVIEWS IN BIOMEDICAL ENGINEERING
♻ ☆ Improving Zero-Shot Object-Level Change Detection by Incorporating Visual Correspondence
Detecting object-level changes between two images across possibly different views is a core task in many applications that involve visual inspection or camera surveillance. Existing change-detection approaches suffer from three major limitations: (1) lack of evaluation on image pairs that contain no changes, leading to unreported false positive rates; (2) lack of correspondences (i.e., localizing the regions before and after a change); and (3) poor zero-shot generalization across different domains. To address these issues, we introduce a novel method that leverages change correspondences (a) during training to improve change detection accuracy, and (b) at test time, to minimize false positives. That is, we harness the supervision labels of where an object is added or removed to supervise change detectors, improving their accuracy over previous work by a large margin. Our work is also the first to predict correspondences between pairs of detected changes using estimated homography and the Hungarian algorithm. Our model demonstrates superior performance over existing methods, achieving state-of-the-art results in change detection and change correspondence accuracy across both in-distribution and zero-shot benchmarks.
♻ ☆ MECD+: Unlocking Event-Level Causal Graph Discovery for Video Reasoning NeurIPS 2024
Video causal reasoning aims to achieve a high-level understanding of videos from a causal perspective. However, it exhibits limitations in its scope, primarily executed in a question-answering paradigm and focusing on brief video segments containing isolated events and basic causal relations, lacking comprehensive and structured causality analysis for videos with multiple interconnected events. To fill this gap, we introduce a new task and dataset, Multi-Event Causal Discovery (MECD). It aims to uncover the causal relations between events distributed chronologically across long videos. Given visual segments and textual descriptions of events, MECD identifies the causal associations between these events to derive a comprehensive and structured event-level video causal graph explaining why and how the result event occurred. To address the challenges of MECD, we devise a novel framework inspired by the Granger Causality method, incorporating an efficient mask-based event prediction model to perform an Event Granger Test. It estimates causality by comparing the predicted result event when premise events are masked versus unmasked. Furthermore, we integrate causal inference techniques such as front-door adjustment and counterfactual inference to mitigate challenges in MECD like causality confounding and illusory causality. Additionally, context chain reasoning is introduced to conduct more robust and generalized reasoning. Experiments validate the effectiveness of our framework in reasoning complete causal relations, outperforming GPT-4o and VideoChat2 by 5.77% and 2.70%, respectively. Further experiments demonstrate that causal relation graphs can also contribute to downstream video understanding tasks such as video question answering and video event prediction.
comment: IEEE TPAMI Submission. continuous work of arXiv:2409.17647 (NeurIPS 2024)
♻ ☆ VITA-1.5: Towards GPT-4o Level Real-Time Vision and Speech Interaction
Recent Multimodal Large Language Models (MLLMs) have typically focused on integrating visual and textual modalities, with less emphasis placed on the role of speech in enhancing interaction. However, speech plays a crucial role in multimodal dialogue systems, and implementing high-performance in both vision and speech tasks remains a significant challenge due to the fundamental modality differences. In this paper, we propose a carefully designed multi-stage training methodology that progressively trains LLM to understand both visual and speech information, ultimately enabling fluent vision and speech interaction. Our approach not only preserves strong vision-language capacity, but also enables efficient speech-to-speech dialogue capabilities without separate ASR and TTS modules, significantly accelerating multimodal end-to-end response speed. By comparing our method against state-of-the-art counterparts across benchmarks for image, video, and speech tasks, we demonstrate that our model is equipped with both strong visual and speech capabilities, making near real-time vision and speech interaction.
comment: https://github.com/VITA-MLLM/VITA
♻ ☆ Bayesian Low-Rank LeArning (Bella): A Practical Approach to Bayesian Neural Networks AAAI'2025
Computational complexity of Bayesian learning is impeding its adoption in practical, large-scale tasks. Despite demonstrations of significant merits such as improved robustness and resilience to unseen or out-of-distribution inputs over their non- Bayesian counterparts, their practical use has faded to near insignificance. In this study, we introduce an innovative framework to mitigate the computational burden of Bayesian neural networks (BNNs). Our approach follows the principle of Bayesian techniques based on deep ensembles, but significantly reduces their cost via multiple low-rank perturbations of parameters arising from a pre-trained neural network. Both vanilla version of ensembles as well as more sophisticated schemes such as Bayesian learning with Stein Variational Gradient Descent (SVGD), previously deemed impractical for large models, can be seamlessly implemented within the proposed framework, called Bayesian Low-Rank LeArning (Bella). In a nutshell, i) Bella achieves a dramatic reduction in the number of trainable parameters required to approximate a Bayesian posterior; and ii) it not only maintains, but in some instances, surpasses the performance of conventional Bayesian learning methods and non-Bayesian baselines. Our results with large-scale tasks such as ImageNet, CAMELYON17, DomainNet, VQA with CLIP, LLaVA demonstrate the effectiveness and versatility of Bella in building highly scalable and practical Bayesian deep models for real-world applications.
comment: This paper is accepted in AAAI'2025
♻ ☆ Latent Space Characterization of Autoencoder Variants
Understanding the latent spaces learned by deep learning models is crucial in exploring how they represent and generate complex data. Autoencoders (AEs) have played a key role in the area of representation learning, with numerous regularization techniques and training principles developed not only to enhance their ability to learn compact and robust representations, but also to reveal how different architectures influence the structure and smoothness of the lower-dimensional non-linear manifold. We strive to characterize the structure of the latent spaces learned by different autoencoders including convolutional autoencoders (CAEs), denoising autoencoders (DAEs), and variational autoencoders (VAEs) and how they change with the perturbations in the input. By characterizing the matrix manifolds corresponding to the latent spaces, we provide an explanation for the well-known observation that the latent spaces of CAE and DAE form non-smooth manifolds, while that of VAE forms a smooth manifold. We also map the points of the matrix manifold to a Hilbert space using distance preserving transforms and provide an alternate view in terms of the subspaces generated in the Hilbert space as a function of the distortion in the input. The results show that the latent manifolds of CAE and DAE are stratified with each stratum being a smooth product manifold, while the manifold of VAE is a smooth product manifold of two symmetric positive definite matrices and a symmetric positive semi-definite matrix.
comment: 9 pages, 6 figures, and 1 table
♻ ☆ STROOBnet Optimization via GPU-Accelerated Proximal Recurrence Strategies
Spatiotemporal networks' observational capabilities are crucial for accurate data gathering and informed decisions across multiple sectors. This study focuses on the Spatiotemporal Ranged Observer-Observable Bipartite Network (STROOBnet), linking observational nodes (e.g., surveillance cameras) to events within defined geographical regions, enabling efficient monitoring. Using data from Real-Time Crime Camera (RTCC) systems and Calls for Service (CFS) in New Orleans, where RTCC combats rising crime amidst reduced police presence, we address the network's initial observational imbalances. Aiming for uniform observational efficacy, we propose the Proximal Recurrence approach. It outperformed traditional clustering methods like k-means and DBSCAN by offering holistic event frequency and spatial consideration, enhancing observational coverage.
comment: 10 pages, 17 figures, 2023 IEEE International Conference on Big Data (BigData)
♻ ☆ Enhancing Few-Shot Image Classification through Learnable Multi-Scale Embedding and Attention Mechanisms
In the context of few-shot classification, the goal is to train a classifier using a limited number of samples while maintaining satisfactory performance. However, traditional metric-based methods exhibit certain limitations in achieving this objective. These methods typically rely on a single distance value between the query feature and support feature, thereby overlooking the contribution of shallow features. To overcome this challenge, we propose a novel approach in this paper. Our approach involves utilizing a multi-output embedding network that maps samples into distinct feature spaces. The proposed method extracts feature vectors at different stages, enabling the model to capture both global and abstract features. By utilizing these diverse feature spaces, our model enhances its performance. Moreover, employing a self-attention mechanism improves the refinement of features at each stage, leading to even more robust representations and improved overall performance. Furthermore, assigning learnable weights to each stage significantly improved performance and results. We conducted comprehensive evaluations on the MiniImageNet and FC100 datasets, specifically in the 5-way 1-shot and 5-way 5-shot scenarios. Additionally, we performed cross-domain tasks across eight benchmark datasets, achieving high accuracy in the testing domains. These evaluations demonstrate the efficacy of our proposed method in comparison to state-of-the-art approaches. https://github.com/FatemehAskari/MSENet
♻ ☆ A Multi-Modal Approach for Face Anti-Spoofing in Non-Calibrated Systems using Disparity Maps
Face recognition technologies are increasingly used in various applications, yet they are vulnerable to face spoofing attacks. These spoofing attacks often involve unique 3D structures, such as printed papers or mobile device screens. Although stereo-depth cameras can detect such attacks effectively, their high-cost limits their widespread adoption. Conversely, two-sensor systems without extrinsic calibration offer a cost-effective alternative but are unable to calculate depth using stereo techniques. In this work, we propose a method to overcome this challenge by leveraging facial attributes to derive disparity information and estimate relative depth for anti-spoofing purposes, using non-calibrated systems. We introduce a multi-modal anti-spoofing model, coined Disparity Model, that incorporates created disparity maps as a third modality alongside the two original sensor modalities. We demonstrate the effectiveness of the Disparity Model in countering various spoof attacks using a comprehensive dataset collected from the Intel RealSense ID Solution F455. Our method outperformed existing methods in the literature, achieving an Equal Error Rate (EER) of 1.71% and a False Negative Rate (FNR) of 2.77% at a False Positive Rate (FPR) of 1%. These errors are lower by 2.45% and 7.94% than the errors of the best comparison method, respectively. Additionally, we introduce a model ensemble that addresses 3D spoof attacks as well, achieving an EER of 2.04% and an FNR of 3.83% at an FPR of 1%. Overall, our work provides a state-of-the-art solution for the challenging task of anti-spoofing in non-calibrated systems that lack depth information.
♻ ☆ Evaluating alignment between humans and neural network representations in image-based learning tasks
Humans represent scenes and objects in rich feature spaces, carrying information that allows us to generalise about category memberships and abstract functions with few examples. What determines whether a neural network model generalises like a human? We tested how well the representations of $86$ pretrained neural network models mapped to human learning trajectories across two tasks where humans had to learn continuous relationships and categories of natural images. In these tasks, both human participants and neural networks successfully identified the relevant stimulus features within a few trials, demonstrating effective generalisation. We found that while training dataset size was a core determinant of alignment with human choices, contrastive training with multi-modal data (text and imagery) was a common feature of currently publicly available models that predicted human generalisation. Intrinsic dimensionality of representations had different effects on alignment for different model types. Lastly, we tested three sets of human-aligned representations and found no consistent improvements in predictive accuracy compared to the baselines. In conclusion, pretrained neural networks can serve to extract representations for cognitive models, as they appear to capture some fundamental aspects of cognition that are transferable across tasks. Both our paradigms and modelling approach offer a novel way to quantify alignment between neural networks and humans and extend cognitive science into more naturalistic domains.
♻ ☆ Instruction-Guided Fusion of Multi-Layer Visual Features in Large Vision-Language Models
Large Vision-Language Models (LVLMs) have achieved significant success in multimodal tasks by combining pre-trained vision encoders and large language models. However, current LVLMs mainly rely on features from the final layers of the vision encoder, neglecting complementary information in shallower layers. While recent methods have explored multi-layer features, they are often task-agnostic. We investigate the contributions of visual features from different encoder layers across 18 benchmarks and 6 task categories. Our results show that multi-layer features provide complementary strengths with varying task dependencies, and uniform fusion performs suboptimally. Based on these findings, we propose an instruction-guided vision aggregator that dynamically integrates multi-layer features based on textual instructions, without increasing the number of visual tokens. Extensive evaluations show superior performance, and analysis reveals the dominance of mid-to-high-level features in semantic tasks and the critical role of low-level features in fine-grained perception. This work provides valuable insights into the adaptive use of hierarchical visual features in LVLMs, advancing more flexible multimodal systems.
♻ ☆ Diffusion Models in Vision: A Survey
Denoising diffusion models represent a recent emerging topic in computer vision, demonstrating remarkable results in the area of generative modeling. A diffusion model is a deep generative model that is based on two stages, a forward diffusion stage and a reverse diffusion stage. In the forward diffusion stage, the input data is gradually perturbed over several steps by adding Gaussian noise. In the reverse stage, a model is tasked at recovering the original input data by learning to gradually reverse the diffusion process, step by step. Diffusion models are widely appreciated for the quality and diversity of the generated samples, despite their known computational burdens, i.e. low speeds due to the high number of steps involved during sampling. In this survey, we provide a comprehensive review of articles on denoising diffusion models applied in vision, comprising both theoretical and practical contributions in the field. First, we identify and present three generic diffusion modeling frameworks, which are based on denoising diffusion probabilistic models, noise conditioned score networks, and stochastic differential equations. We further discuss the relations between diffusion models and other deep generative models, including variational auto-encoders, generative adversarial networks, energy-based models, autoregressive models and normalizing flows. Then, we introduce a multi-perspective categorization of diffusion models applied in computer vision. Finally, we illustrate the current limitations of diffusion models and envision some interesting directions for future research.
comment: Accepted in IEEE Transactions on Pattern Analysis and Machine Intelligence. 25 pages, 3 figures
♻ ☆ DriveLM: Driving with Graph Visual Question Answering ECCV 2024
We study how vision-language models (VLMs) trained on web-scale data can be integrated into end-to-end driving systems to boost generalization and enable interactivity with human users. While recent approaches adapt VLMs to driving via single-round visual question answering (VQA), human drivers reason about decisions in multiple steps. Starting from the localization of key objects, humans estimate object interactions before taking actions. The key insight is that with our proposed task, Graph VQA, where we model graph-structured reasoning through perception, prediction and planning question-answer pairs, we obtain a suitable proxy task to mimic the human reasoning process. We instantiate datasets (DriveLM-Data) built upon nuScenes and CARLA, and propose a VLM-based baseline approach (DriveLM-Agent) for jointly performing Graph VQA and end-to-end driving. The experiments demonstrate that Graph VQA provides a simple, principled framework for reasoning about a driving scene, and DriveLM-Data provides a challenging benchmark for this task. Our DriveLM-Agent baseline performs end-to-end autonomous driving competitively in comparison to state-of-the-art driving-specific architectures. Notably, its benefits are pronounced when it is evaluated zero-shot on unseen objects or sensor configurations. We hope this work can be the starting point to shed new light on how to apply VLMs for autonomous driving. To facilitate future research, all code, data, and models are available to the public.
comment: Accepted to ECCV 2024 as Oral paper
♻ ☆ Towards an End-to-End (E2E) Adversarial Learning and Application in the Physical World
The traditional learning process of patch-based adversarial attacks, conducted in the digital domain and then applied in the physical domain (e.g., via printed stickers), may suffer from reduced performance due to adversarial patches' limited transferability from the digital domain to the physical domain. Given that previous studies have considered using projectors to apply adversarial attacks, we raise the following question: can adversarial learning (i.e., patch generation) be performed entirely in the physical domain with a projector? In this work, we propose the Physical-domain Adversarial Patch Learning Augmentation (PAPLA) framework, a novel end-to-end (E2E) framework that converts adversarial learning from the digital domain to the physical domain using a projector. We evaluate PAPLA across multiple scenarios, including controlled laboratory settings and realistic outdoor environments, demonstrating its ability to ensure attack success compared to conventional digital learning-physical application (DL-PA) methods. We also analyze the impact of environmental factors, such as projection surface color, projector strength, ambient light, distance, and angle of the target object relative to the camera, on the effectiveness of projected patches. Finally, we demonstrate the feasibility of the attack against a parked car and a stop sign in a real-world outdoor environment. Our results show that under specific conditions, E2E adversarial learning in the physical domain eliminates the transferability issue and ensures evasion by object detectors. Finally, we provide insights into the challenges and opportunities of applying adversarial learning in the physical domain and explain where such an approach is more effective than using a sticker.
♻ ☆ TextureCrop: Enhancing Synthetic Image Detection through Texture-based Cropping
Generative AI technologies produce increasingly realistic imagery, which, despite its potential for creative applications, can also be misused to produce misleading and harmful content. This renders Synthetic Image Detection (SID) methods essential for identifying AI-generated content online. State-of-the-art SID methods typically resize or center-crop input images due to architectural or computational constraints, which hampers the detection of artifacts that appear in high-resolution images. To address this limitation, we propose TextureCrop, an image pre-processing component that can be plugged in any pre-trained SID model to improve its performance. By focusing on high-frequency image parts where generative artifacts are prevalent, TextureCrop enhances SID performance with manageable memory requirements. Experimental results demonstrate a consistent improvement in AUC across various detectors by 6.1% compared to center cropping and by 15% compared to resizing, across high-resolution images from the Forensynths, Synthbuster and TWIGMA datasets. Code available at https : //github.com/mever-team/texture-crop.
comment: 10 pages, 7 images
♻ ☆ IOR: Inversed Objects Replay for Incremental Object Detection
Existing Incremental Object Detection (IOD) methods partially alleviate catastrophic forgetting when incrementally detecting new objects in real-world scenarios. However, many of these methods rely on the assumption that unlabeled old-class objects may co-occur with labeled new-class objects in the incremental data. When unlabeled old-class objects are absent, the performance of existing methods tends to degrade. The absence can be mitigated by generating old-class samples, but it incurs high costs. This paper argues that previous generation-based IOD suffers from redundancy, both in the use of generative models, which require additional training and storage, and in the overproduction of generated samples, many of which do not contribute significantly to performance improvements. To eliminate the redundancy, we propose Inversed Objects Replay (IOR). Specifically, we generate old-class samples by inversing the original detectors, thus eliminating the necessity of training and storing additional generative models. We propose augmented replay to reuse the objects in generated samples, reducing redundant generations. Moreover, we propose high-value knowledge distillation focusing on the positions of old-class objects overwhelmed by the background, which transfers the knowledge to the incremental detector. Extensive experiments conducted on MS COCO 2017 demonstrate that our method can efficiently improve detection performance in IOD scenarios with the absence of old-class objects.
♻ ☆ Skinned Motion Retargeting with Dense Geometric Interaction Perception NeurIPS 2024
Capturing and maintaining geometric interactions among different body parts is crucial for successful motion retargeting in skinned characters. Existing approaches often overlook body geometries or add a geometry correction stage after skeletal motion retargeting. This results in conflicts between skeleton interaction and geometry correction, leading to issues such as jittery, interpenetration, and contact mismatches. To address these challenges, we introduce a new retargeting framework, MeshRet, which directly models the dense geometric interactions in motion retargeting. Initially, we establish dense mesh correspondences between characters using semantically consistent sensors (SCS), effective across diverse mesh topologies. Subsequently, we develop a novel spatio-temporal representation called the dense mesh interaction (DMI) field. This field, a collection of interacting SCS feature vectors, skillfully captures both contact and non-contact interactions between body geometries. By aligning the DMI field during retargeting, MeshRet not only preserves motion semantics but also prevents self-interpenetration and ensures contact preservation. Extensive experiments on the public Mixamo dataset and our newly-collected ScanRet dataset demonstrate that MeshRet achieves state-of-the-art performance. Code available at https://github.com/abcyzj/MeshRet.
comment: NeurIPS 2024 Spotlight
♻ ☆ reBEN: Refined BigEarthNet Dataset for Remote Sensing Image Analysis
This paper presents refined BigEarthNet (reBEN) that is a large-scale, multi-modal remote sensing dataset constructed to support deep learning (DL) studies for remote sensing image analysis. The reBEN dataset consists of 549,488 pairs of Sentinel-1 and Sentinel-2 image patches. To construct reBEN, we initially consider the Sentinel-1 and Sentinel-2 tiles used to construct the BigEarthNet dataset and then divide them into patches of size 1200 m x 1200 m. We apply atmospheric correction to the Sentinel-2 patches using the latest version of the sen2cor tool, resulting in higher-quality patches compared to those present in BigEarthNet. Each patch is then associated with a pixel-level reference map and scene-level multi-labels. This makes reBEN suitable for pixel- and scene-based learning tasks. The labels are derived from the most recent CORINE Land Cover (CLC) map of 2018 by utilizing the 19-class nomenclature as in BigEarthNet. The use of the most recent CLC map results in overcoming the label noise present in BigEarthNet. Furthermore, we introduce a new geographical-based split assignment algorithm that significantly reduces the spatial correlation among the train, validation, and test sets with respect to those present in BigEarthNet. This increases the reliability of the evaluation of DL models. To minimize the DL model training time, we introduce software tools that convert the reBEN dataset into a DL-optimized data format. In our experiments, we show the potential of reBEN for multi-modal multi-label image classification problems by considering several state-of-the-art DL models. The pre-trained model weights, associated code, and complete dataset are available at https://bigearth.net.
♻ ☆ DehazeGS: Seeing Through Fog with 3D Gaussian Splatting
Current novel view synthesis tasks primarily rely on high-quality and clear images. However, in foggy scenes, scattering and attenuation can significantly degrade the reconstruction and rendering quality. Although NeRF-based dehazing reconstruction algorithms have been developed, their use of deep fully connected neural networks and per-ray sampling strategies leads to high computational costs. Moreover, NeRF's implicit representation struggles to recover fine details from hazy scenes. In contrast, recent advancements in 3D Gaussian Splatting achieve high-quality 3D scene reconstruction by explicitly modeling point clouds into 3D Gaussians. In this paper, we propose leveraging the explicit Gaussian representation to explain the foggy image formation process through a physically accurate forward rendering process. We introduce DehazeGS, a method capable of decomposing and rendering a fog-free background from participating media using only muti-view foggy images as input. We model the transmission within each Gaussian distribution to simulate the formation of fog. During this process, we jointly learn the atmospheric light and scattering coefficient while optimizing the Gaussian representation of the hazy scene. In the inference stage, we eliminate the effects of scattering and attenuation on the Gaussians and directly project them onto a 2D plane to obtain a clear view. Experiments on both synthetic and real-world foggy datasets demonstrate that DehazeGS achieves state-of-the-art performance in terms of both rendering quality and computational efficiency. visualizations are available at https://dehazegs.github.io/
comment: 9 pages,4 figures
♻ ☆ StructSR: Refuse Spurious Details in Real-World Image Super-Resolution
Diffusion-based models have shown great promise in real-world image super-resolution (Real-ISR), but often generate content with structural errors and spurious texture details due to the empirical priors and illusions of these models. To address this issue, we introduce StructSR, a simple, effective, and plug-and-play method that enhances structural fidelity and suppresses spurious details for diffusion-based Real-ISR. StructSR operates without the need for additional fine-tuning, external model priors, or high-level semantic knowledge. At its core is the Structure-Aware Screening (SAS) mechanism, which identifies the image with the highest structural similarity to the low-resolution (LR) input in the early inference stage, allowing us to leverage it as a historical structure knowledge to suppress the generation of spurious details. By intervening in the diffusion inference process, StructSR seamlessly integrates with existing diffusion-based Real-ISR models. Our experimental results demonstrate that StructSR significantly improves the fidelity of structure and texture, improving the PSNR and SSIM metrics by an average of 5.27% and 9.36% on a synthetic dataset (DIV2K-Val) and 4.13% and 8.64% on two real-world datasets (RealSR and DRealSR) when integrated with four state-of-the-art diffusion-based Real-ISR methods.
♻ ☆ Direct Unlearning Optimization for Robust and Safe Text-to-Image Models NeurIPS 2024
Recent advancements in text-to-image (T2I) models have unlocked a wide range of applications but also present significant risks, particularly in their potential to generate unsafe content. To mitigate this issue, researchers have developed unlearning techniques to remove the model's ability to generate potentially harmful content. However, these methods are easily bypassed by adversarial attacks, making them unreliable for ensuring the safety of generated images. In this paper, we propose Direct Unlearning Optimization (DUO), a novel framework for removing Not Safe For Work (NSFW) content from T2I models while preserving their performance on unrelated topics. DUO employs a preference optimization approach using curated paired image data, ensuring that the model learns to remove unsafe visual concepts while retaining unrelated features. Furthermore, we introduce an output-preserving regularization term to maintain the model's generative capabilities on safe content. Extensive experiments demonstrate that DUO can robustly defend against various state-of-the-art red teaming methods without significant performance degradation on unrelated topics, as measured by FID and CLIP scores. Our work contributes to the development of safer and more reliable T2I models, paving the way for their responsible deployment in both closed-source and open-source scenarios.
comment: This paper has been accepted for NeurIPS 2024
♻ ☆ Geometric Distortion Guided Transformer for Omnidirectional Image Super-Resolution
As virtual and augmented reality applications gain popularity, omnidirectional image (ODI) super-resolution has become increasingly important. Unlike 2D plain images that are formed on a plane, ODIs are projected onto spherical surfaces. Applying established image super-resolution methods to ODIs, therefore, requires performing equirectangular projection (ERP) to map the ODIs onto a plane. ODI super-resolution needs to take into account geometric distortion resulting from ERP. However, without considering such geometric distortion of ERP images, previous deep-learning-based methods only utilize a limited range of pixels and may easily miss self-similar textures for reconstruction. In this paper, we introduce a novel Geometric Distortion Guided Transformer for Omnidirectional image Super-Resolution (GDGT-OSR). Specifically, a distortion modulated rectangle-window self-attention mechanism, integrated with deformable self-attention, is proposed to better perceive the distortion and thus involve more self-similar textures. Distortion modulation is achieved through a newly devised distortion guidance generator that produces guidance by exploiting the variability of distortion across latitudes. Furthermore, we propose a dynamic feature aggregation scheme to adaptively fuse the features from different self-attention modules. We present extensive experimental results on public datasets and show that the new GDGT-OSR outperforms methods in existing literature.
comment: 13 pages, 12 figures, journal
♻ ☆ iFADIT: Invertible Face Anonymization via Disentangled Identity Transform
Face anonymization aims to conceal the visual identity of a face to safeguard the individual's privacy. Traditional methods like blurring and pixelation can largely remove identifying features, but these techniques significantly degrade image quality and are vulnerable to deep reconstruction attacks. Generative models have emerged as a promising solution for anonymizing faces while preserving a natural appearance. However, many still face limitations in visual quality and often overlook the potential to recover the original face from the anonymized version, which can be valuable in specific contexts such as image forensics. This paper proposes a novel framework named iFADIT, an acronym for Invertible Face Anonymization via Disentangled Identity Transform. The framework features a disentanglement architecture coupled with a secure flow-based model: the former decouples identity information from non-identifying attributes, while the latter transforms the decoupled identity into an anonymized version in an invertible manner controlled by a secret key. The anonymized face can then be reconstructed based on a pre-trained StyleGAN that ensures high image quality and realistic facial details. Recovery of the original face (aka de-anonymization) is possible upon the availability of the matching secret, by inverting the anonymization process based on the same set of model parameters. Furthermore, a dedicated secret-key mechanism along with a dual-phase training strategy is devised to ensure the desired properties of face anonymization. Qualitative and quantitative experiments demonstrate the superiority of the proposed approach in anonymity, reversibility, security, diversity, and interpretability over competing methods.
♻ ☆ Go-with-the-Flow: Motion-Controllable Video Diffusion Models Using Real-Time Warped Noise
Generative modeling aims to transform random noise into structured outputs. In this work, we enhance video diffusion models by allowing motion control via structured latent noise sampling. This is achieved by just a change in data: we pre-process training videos to yield structured noise. Consequently, our method is agnostic to diffusion model design, requiring no changes to model architectures or training pipelines. Specifically, we propose a novel noise warping algorithm, fast enough to run in real time, that replaces random temporal Gaussianity with correlated warped noise derived from optical flow fields, while preserving the spatial Gaussianity. The efficiency of our algorithm enables us to fine-tune modern video diffusion base models using warped noise with minimal overhead, and provide a one-stop solution for a wide range of user-friendly motion control: local object motion control, global camera movement control, and motion transfer. The harmonization between temporal coherence and spatial Gaussianity in our warped noise leads to effective motion control while maintaining per-frame pixel quality. Extensive experiments and user studies demonstrate the advantages of our method, making it a robust and scalable approach for controlling motion in video diffusion models. Video results are available on our webpage: https://vgenai-netflix-eyeline-research.github.io/Go-with-the-Flow. Source code and model checkpoints are available on GitHub: https://github.com/VGenAI-Netflix-Eyeline-Research/Go-with-the-Flow.
♻ ☆ Point-PRC: A Prompt Learning Based Regulation Framework for Generalizable Point Cloud Analysis NeurIPS 2024
This paper investigates the 3D domain generalization (3DDG) ability of large 3D models based on prevalent prompt learning. Recent works demonstrate the performances of 3D point cloud recognition can be boosted remarkably by parameter-efficient prompt tuning. However, we observe that the improvement on downstream tasks comes at the expense of a severe drop in 3D domain generalization. To resolve this challenge, we present a comprehensive regulation framework that allows the learnable prompts to actively interact with the well-learned general knowledge in large 3D models to maintain good generalization. Specifically, the proposed framework imposes multiple explicit constraints on the prompt learning trajectory by maximizing the mutual agreement between task-specific predictions and task-agnostic knowledge. We design the regulation framework as a plug-and-play module to embed into existing representative large 3D models. Surprisingly, our method not only realizes consistently increasing generalization ability but also enhances task-specific 3D recognition performances across various 3DDG benchmarks by a clear margin. Considering the lack of study and evaluation on 3DDG, we also create three new benchmarks, namely base-to-new, cross-dataset and few-shot generalization benchmarks, to enrich the field and inspire future research. Code and benchmarks are available at \url{https://github.com/auniquesun/Point-PRC}.
comment: 5 figures, 14 tables; accepted by NeurIPS 2024
♻ ☆ CMRxRecon2024: A Multi-Modality, Multi-View K-Space Dataset Boosting Universal Machine Learning for Accelerated Cardiac MRI
Cardiac magnetic resonance imaging (MRI) has emerged as a clinically gold-standard technique for diagnosing cardiac diseases, thanks to its ability to provide diverse information with multiple modalities and anatomical views. Accelerated cardiac MRI is highly expected to achieve time-efficient and patient-friendly imaging, and then advanced image reconstruction approaches are required to recover high-quality, clinically interpretable images from undersampled measurements. However, the lack of publicly available cardiac MRI k-space dataset in terms of both quantity and diversity has severely hindered substantial technological progress, particularly for data-driven artificial intelligence. Here, we provide a standardized, diverse, and high-quality CMRxRecon2024 dataset to facilitate the technical development, fair evaluation, and clinical transfer of cardiac MRI reconstruction approaches, towards promoting the universal frameworks that enable fast and robust reconstructions across different cardiac MRI protocols in clinical practice. To the best of our knowledge, the CMRxRecon2024 dataset is the largest and most protocal-diverse publicly available cardiac k-space dataset. It is acquired from 330 healthy volunteers, covering commonly used modalities, anatomical views, and acquisition trajectories in clinical cardiac MRI workflows. Besides, an open platform with tutorials, benchmarks, and data processing tools is provided to facilitate data usage, advanced method development, and fair performance evaluation.
comment: 23 pages, 3 figures, 2 tables
♻ ☆ VLG-CBM: Training Concept Bottleneck Models with Vision-Language Guidance NeurIPS 2024
Concept Bottleneck Models (CBMs) provide interpretable prediction by introducing an intermediate Concept Bottleneck Layer (CBL), which encodes human-understandable concepts to explain models' decision. Recent works proposed to utilize Large Language Models and pre-trained Vision-Language Models to automate the training of CBMs, making it more scalable and automated. However, existing approaches still fall short in two aspects: First, the concepts predicted by CBL often mismatch the input image, raising doubts about the faithfulness of interpretation. Second, it has been shown that concept values encode unintended information: even a set of random concepts could achieve comparable test accuracy to state-of-the-art CBMs. To address these critical limitations, in this work, we propose a novel framework called Vision-Language-Guided Concept Bottleneck Model (VLG-CBM) to enable faithful interpretability with the benefits of boosted performance. Our method leverages off-the-shelf open-domain grounded object detectors to provide visually grounded concept annotation, which largely enhances the faithfulness of concept prediction while further improving the model performance. In addition, we propose a new metric called Number of Effective Concepts (NEC) to control the information leakage and provide better interpretability. Extensive evaluations across five standard benchmarks show that our method, VLG-CBM, outperforms existing methods by at least 4.27% and up to 51.09% on Accuracy at NEC=5 (denoted as ANEC-5), and by at least 0.45% and up to 29.78% on average accuracy (denoted as ANEC-avg), while preserving both faithfulness and interpretability of the learned concepts as demonstrated in extensive experiments.
comment: Appeared at NeurIPS 2024
♻ ☆ Synthesizing Forestry Images Conditioned on Plant Phenotype Using a Generative Adversarial Network
Plant phenology and phenotype prediction using remote sensing data are increasingly gaining attention within the plant science community as a promising approach to enhance agricultural productivity. This work focuses on generating synthetic forestry images that satisfy certain phenotypic attributes, viz. canopy greenness. We harness a Generative Adversarial Network (GAN) to synthesize biologically plausible and phenotypically stable forestry images conditioned on the greenness of vegetation (a continuous attribute) over a specific region of interest, describing a particular vegetation type in a mixed forest. The training data is based on the automated digital camera imagery provided by the National Ecological Observatory Network (NEON) and processed by the PhenoCam Network. Our method helps render the appearance of forest sites specific to a greenness value. The synthetic images are subsequently utilized to predict another phenotypic attribute, viz., redness of plants. The quality of the synthetic images is assessed using the Structural SIMilarity (SSIM) index and Fr\'echet Inception Distance (FID). Further, the greenness and redness indices of the synthetic images are compared against those of the original images using Root Mean Squared Percentage Error (RMSPE) to evaluate their accuracy and integrity. The generalizability and scalability of our proposed GAN model are established by effectively transforming it to generate synthetic images for other forest sites and vegetation types. From a broader perspective, this approach could be leveraged to visualize forestry based on different phenotypic attributes in the context of various environmental parameters.
comment: Accepted to Pattern Recognition journal
♻ ☆ BRIGHT-VO: Brightness-Guided Hybrid Transformer for Visual Odometry with Multi-modality Refinement Module
Visual odometry (VO) plays a crucial role in autonomous driving, robotic navigation, and other related tasks by estimating the position and orientation of a camera based on visual input. Significant progress has been made in data-driven VO methods, particularly those leveraging deep learning techniques to extract image features and estimate camera poses. However, these methods often struggle in low-light conditions because of the reduced visibility of features and the increased difficulty of matching keypoints. To address this limitation, we introduce BrightVO, a novel VO model based on Transformer architecture, which not only performs front-end visual feature extraction, but also incorporates a multi-modality refinement module in the back-end that integrates Inertial Measurement Unit (IMU) data. Using pose graph optimization, this module iteratively refines pose estimates to reduce errors and improve both accuracy and robustness. Furthermore, we create a synthetic low-light dataset, KiC4R, which includes a variety of lighting conditions to facilitate the training and evaluation of VO frameworks in challenging environments. Experimental results demonstrate that BrightVO achieves state-of-the-art performance on both the KiC4R dataset and the KITTI benchmarks. Specifically, it provides an average improvement of 20% in pose estimation accuracy in normal outdoor environments and 259% in low-light conditions, outperforming existing methods. For widespread use and further development, the research work is fully open-source at https://github.com/Anastasiawd/BrightVO.
comment: We have identified significant issues in the methodology and data analysis that impact the validity of our conclusions
♻ ☆ A General Framework for Inference-time Scaling and Steering of Diffusion Models
Diffusion models produce impressive results in modalities ranging from images and video to protein design and text. However, generating samples with user-specified properties remains a challenge. Recent research proposes fine-tuning models to maximize rewards that capture desired properties, but these methods require expensive training and are prone to mode collapse. In this work, we propose Feynman Kac (FK) steering, an inference-time framework for steering diffusion models with reward functions. FK steering works by sampling a system of multiple interacting diffusion processes, called particles, and resampling particles at intermediate steps based on scores computed using functions called potentials. Potentials are defined using rewards for intermediate states and are selected such that a high value indicates that the particle will yield a high-reward sample. We explore various choices of potentials, intermediate rewards, and samplers. We evaluate FK steering on text-to-image and text diffusion models. For steering text-to-image models with a human preference reward, we find that FK steering a 0.8B parameter model outperforms a 2.6B parameter fine-tuned model on prompt fidelity, with faster sampling and no training. For steering text diffusion models with rewards for text quality and specific text attributes, we find that FK steering generates lower perplexity, more linguistically acceptable outputs and enables gradient-free control of attributes like toxicity. Our results demonstrate that inference-time scaling and steering of diffusion models, even with off-the-shelf rewards, can provide significant sample quality gains and controllability benefits. Code is available at https://github.com/zacharyhorvitz/Fk-Diffusion-Steering .
♻ ☆ DiffMesh: A Motion-aware Diffusion Framework for Human Mesh Recovery from Videos WACV 2025
Human mesh recovery (HMR) provides rich human body information for various real-world applications. While image-based HMR methods have achieved impressive results, they often struggle to recover humans in dynamic scenarios, leading to temporal inconsistencies and non-smooth 3D motion predictions due to the absence of human motion. In contrast, video-based approaches leverage temporal information to mitigate this issue. In this paper, we present DiffMesh, an innovative motion-aware Diffusion-like framework for video-based HMR. DiffMesh establishes a bridge between diffusion models and human motion, efficiently generating accurate and smooth output mesh sequences by incorporating human motion within the forward process and reverse process in the diffusion model. Extensive experiments are conducted on the widely used datasets (Human3.6M \cite{h36m_pami} and 3DPW \cite{pw3d2018}), which demonstrate the effectiveness and efficiency of our DiffMesh. Visual comparisons in real-world scenarios further highlight DiffMesh's suitability for practical applications.
comment: WACV 2025
♻ ☆ Towards Balanced Continual Multi-Modal Learning in Human Pose Estimation
3D human pose estimation (3D HPE) has emerged as a prominent research topic, particularly in the realm of RGB-based methods. However, RGB images are susceptible to limitations such as sensitivity to lighting conditions and potential user discomfort. Consequently, multi-modal sensing, which leverages non-intrusive sensors, is gaining increasing attention. Nevertheless, multi-modal 3D HPE still faces challenges, including modality imbalance and the imperative for continual learning. In this work, we introduce a novel balanced continual multi-modal learning method for 3D HPE, which harnesses the power of RGB, LiDAR, mmWave, and WiFi. Specifically, we propose a Shapley value-based contribution algorithm to quantify the contribution of each modality and identify modality imbalance. To address this imbalance, we employ a re-learning strategy. Furthermore, recognizing that raw data is prone to noise contamination, we develop a novel denoising continual learning approach. This approach incorporates a noise identification and separation module to mitigate the adverse effects of noise and collaborates with the balanced learning strategy to enhance optimization. Additionally, an adaptive EWC mechanism is employed to alleviate catastrophic forgetting. We conduct extensive experiments on the widely-adopted multi-modal dataset, MM-Fi, which demonstrate the superiority of our approach in boosting 3D pose estimation and mitigating catastrophic forgetting in complex scenarios. We will release our codes.
♻ ☆ Collaboration in Immersive Environments: Challenges and Solutions
Virtual Reality (VR) and Augmented Reality (AR) tools have been applied in all engineering fields in order to avoid the use of physical prototypes, to train in high-risk situations, and to interpret real or simulated results. In order to complete a shared task or assign tasks to the agents in such immersive environments, collaboration or Shared Cooperative Activities are a necessity. Collaboration in immersive environments is an emerging field of research that aims to study and enhance the ways in which people interact and work together in Virtual and Augmented Reality settings. Collaboration in immersive environments is a complex process that involves different factors such as communication, coordination, and social presence. This paper provides an overview of the current state of research on collaboration in immersive environments. It discusses the different types of immersive environments, including VR and AR, and the different forms of collaboration that can occur in these environments. The paper also highlights the challenges and limitations of collaboration in immersive environments, such as the lack of physical cues, cost and usability and the need for further research in this area. Overall, collaboration in immersive environments is a promising field with a wide range of potential applications, from education to industry, and it can benefit both individuals and groups by enhancing their ability to work together effectively.
comment: Added new references in Networking section
♻ ☆ Rethinking Pre-Trained Feature Extractor Selection in Multiple Instance Learning for Whole Slide Image Classification
Multiple instance learning (MIL) has become a preferred method for gigapixel whole slide image (WSI) classification without requiring patch-level annotations. Current MIL research primarily relies on embedding-based approaches, which extract patch features using a pre-trained feature extractor and aggregate them for slide-level prediction. Despite the critical role of feature extraction, there is limited guidance on selecting optimal feature extractors to maximize WSI performance. This study addresses this gap by systematically evaluating MIL feature extractors across three dimensions: pre-training dataset, backbone model, and pre-training method. Extensive experiments were conducted on two public WSI datasets (TCGA-NSCLC and Camelyon16) using four state-of-the-art (SOTA) MIL models. Our findings reveal that: 1) selecting a robust self-supervised learning (SSL) method has a greater impact on performance than relying solely on an in-domain pre-training dataset; 2) prioritizing Transformer-based backbones with deeper architectures over CNN-based models; and 3) using larger, more diverse pre-training datasets significantly enhances classification outcomes. We hope that these insights can provide practical guidance for optimizing WSI classification and explain the reasons behind the performance advantages of the current SOTA pathology foundation models. Furthermore, this work may inform the development of more effective pathology foundation models. Our code is publicly available at https://github.com/bryanwong17/MIL-Feature-Extractor-Selection
comment: Accepted to IEEE International Symposium on Biomedical Imaging (ISBI) 2025
♻ ☆ PhysMamba: State Space Duality Model for Remote Physiological Measurement
Remote Photoplethysmography (rPPG) enables non-contact physiological signal extraction from facial videos, offering applications in psychological state analysis, medical assistance, and anti-face spoofing. However, challenges such as motion artifacts, lighting variations, and noise limit its real-world applicability. To address these issues, we propose PhysMamba, a novel dual-pathway time-frequency interaction model based on Synergistic State Space Duality (SSSD), which for the first time integrates state space models with attention mechanisms in a dual-branch framework. Combined with a Multi-Scale Query (MQ) mechanism, PhysMamba achieves efficient information exchange and enhanced feature representation, ensuring robustness under noisy and dynamic conditions. Experiments on PURE, UBFC-rPPG, and MMPD datasets demonstrate that PhysMamba outperforms state-of-the-art methods, offering superior accuracy and generalization. This work lays a strong foundation for practical applications in non-contact health monitoring, including real-time remote patient care.
♻ ☆ Enhanced Masked Image Modeling to Avoid Model Collapse on Multi-modal MRI Datasets
Multi-modal magnetic resonance imaging (MRI) provides information of lesions for computer-aided diagnosis from different views. Deep learning algorithms are suitable for identifying specific anatomical structures, segmenting lesions, and classifying diseases. Manual labels are limited due to the high expense, which hinders further improvement of accuracy. Self-supervised learning, particularly masked image modeling (MIM), has shown promise in utilizing unlabeled data. However, we spot model collapse when applying MIM to multi-modal MRI datasets. The performance of downstream tasks does not see any improvement following the collapsed model. To solve model collapse, we analyze and address it in two types: complete collapse and dimensional collapse. We find complete collapse occurs because the collapsed loss value in multi-modal MRI datasets falls below the normally converged loss value. Based on this, the hybrid mask pattern (HMP) masking strategy is introduced to elevate the collapsed loss above the normally converged loss value and avoid complete collapse. Additionally, we reveal that dimensional collapse stems from insufficient feature uniformity in MIM. We mitigate dimensional collapse by introducing the pyramid barlow twins (PBT) module as an explicit regularization method. Overall, we construct the enhanced MIM (E-MIM) with HMP and PBT module to avoid model collapse multi-modal MRI. Experiments are conducted on three multi-modal MRI datasets to validate the effectiveness of our approach in preventing both types of model collapse. By preventing model collapse, the training of the model becomes more stable, resulting in a decent improvement in performance for segmentation and classification tasks. The code is available at https://github.com/LinxuanHan/E-MIM.
comment: This work has been submitted to the lEEE for possible publication. copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ CryoBench: Diverse and challenging datasets for the heterogeneity problem in cryo-EM NeurIPS 2024
Cryo-electron microscopy (cryo-EM) is a powerful technique for determining high-resolution 3D biomolecular structures from imaging data. Its unique ability to capture structural variability has spurred the development of heterogeneous reconstruction algorithms that can infer distributions of 3D structures from noisy, unlabeled imaging data. Despite the growing number of advanced methods, progress in the field is hindered by the lack of standardized benchmarks with ground truth information and reliable validation metrics. Here, we introduce CryoBench, a suite of datasets, metrics, and benchmarks for heterogeneous reconstruction in cryo-EM. CryoBench includes five datasets representing different sources of heterogeneity and degrees of difficulty. These include conformational heterogeneity generated from designed motions of antibody complexes or sampled from a molecular dynamics simulation, as well as compositional heterogeneity from mixtures of ribosome assembly states or 100 common complexes present in cells. We then analyze state-of-the-art heterogeneous reconstruction tools, including neural and non-neural methods, assess their sensitivity to noise, and propose new metrics for quantitative evaluation. We hope that CryoBench will be a foundational resource for accelerating algorithmic development and evaluation in the cryo-EM and machine learning communities. Project page: https://cryobench.cs.princeton.edu.
comment: Accepted by NeurIPS 2024 (Spotlight)
♻ ☆ Swin transformers are robust to distribution and concept drift in endoscopy-based longitudinal rectal cancer assessment SP
Endoscopic images are used at various stages of rectal cancer treatment starting from cancer screening, diagnosis, during treatment to assess response and toxicity from treatments such as colitis, and at follow up to detect new tumor or local regrowth (LR). However, subjective assessment is highly variable and can underestimate the degree of response in some patients, subjecting them to unnecessary surgery, or overestimate response that places patients at risk of disease spread. Advances in deep learning has shown the ability to produce consistent and objective response assessment for endoscopic images. However, methods for detecting cancers, regrowth, and monitoring response during the entire course of patient treatment and follow-up are lacking. This is because, automated diagnosis and rectal cancer response assessment requires methods that are robust to inherent imaging illumination variations and confounding conditions (blood, scope, blurring) present in endoscopy images as well as changes to the normal lumen and tumor during treatment. Hence, a hierarchical shifted window (Swin) transformer was trained to distinguish rectal cancer from normal lumen using endoscopy images. Swin as well as two convolutional (ResNet-50, WideResNet-50), and vision transformer (ViT) models were trained and evaluated on follow-up longitudinal images to detect LR on private dataset as well as on out-of-distribution (OOD) public colonoscopy datasets to detect pre/non-cancerous polyps. Color shifts were applied using optimal transport to simulate distribution shifts. Swin and ResNet models were similarly accurate in the in-distribution dataset. Swin was more accurate than other methods (follow-up: 0.84, OOD: 0.83) even when subject to color shifts (follow-up: 0.83, OOD: 0.87), indicating capability to provide robust performance for longitudinal cancer assessment.
comment: The work has been accepted for publication in 2024 SPIE Medical Imaging conference proceedings
♻ ☆ TakuNet: an Energy-Efficient CNN for Real-Time Inference on Embedded UAV systems in Emergency Response Scenarios WACV
Designing efficient neural networks for embedded devices is a critical challenge, particularly in applications requiring real-time performance, such as aerial imaging with drones and UAVs for emergency responses. In this work, we introduce TakuNet, a novel light-weight architecture which employs techniques such as depth-wise convolutions and an early downsampling stem to reduce computational complexity while maintaining high accuracy. It leverages dense connections for fast convergence during training and uses 16-bit floating-point precision for optimization on embedded hardware accelerators. Experimental evaluation on two public datasets shows that TakuNet achieves near-state-of-the-art accuracy in classifying aerial images of emergency situations, despite its minimal parameter count. Real-world tests on embedded devices, namely Jetson Orin Nano and Raspberry Pi, confirm TakuNet's efficiency, achieving more than 650 fps on the 15W Jetson board, making it suitable for real-time AI processing on resource-constrained platforms and advancing the applicability of drones in emergency scenarios. The code and implementation details are publicly released.
comment: This paper has been accepted at WACVW 2025, which will take place on 28/02/2025. The official conference proceedings have not yet been published at the time of submission to arXiv. The final version of the paper, incorporating any changes based on feedback received during the conference, will be included in the proceedings once they are made available
♻ ☆ AgRegNet: A Deep Regression Network for Flower and Fruit Density Estimation, Localization, and Counting in Orchards
One of the major challenges for the agricultural industry today is the uncertainty in manual labor availability and the associated cost. Automated flower and fruit density estimation, localization, and counting could help streamline harvesting, yield estimation, and crop-load management strategies such as flower and fruitlet thinning. This article proposes a deep regression-based network, AgRegNet, to estimate density, count, and location of flower and fruit in tree fruit canopies without explicit object detection or polygon annotation. Inspired by popular U-Net architecture, AgRegNet is a U-shaped network with an encoder-to-decoder skip connection and modified ConvNeXt-T as an encoder feature extractor. AgRegNet can be trained based on information from point annotation and leverages segmentation information and attention modules (spatial and channel) to highlight relevant flower and fruit features while suppressing non-relevant background features. Experimental evaluation in apple flower and fruit canopy images under an unstructured orchard environment showed that AgRegNet achieved promising accuracy as measured by Structural Similarity Index (SSIM), percentage Mean Absolute Error (pMAE) and mean Average Precision (mAP) to estimate flower and fruit density, count, and centroid location, respectively. Specifically, the SSIM, pMAE, and mAP values for flower images were 0.938, 13.7%, and 0.81, respectively. For fruit images, the corresponding values were 0.910, 5.6%, and 0.93. Since the proposed approach relies on information from point annotation, it is suitable for sparsely and densely located objects. This simplified technique will be highly applicable for growers to accurately estimate yields and decide on optimal chemical and mechanical flower thinning practices.
comment: Published in Computers and Electronics in Agriculture
Computation and Language 66
☆ OmniThink: Expanding Knowledge Boundaries in Machine Writing through Thinking
Machine writing with large language models often relies on retrieval-augmented generation. However, these approaches remain confined within the boundaries of the model's predefined scope, limiting the generation of content with rich information. Specifically, vanilla-retrieved information tends to lack depth, utility, and suffers from redundancy, which negatively impacts the quality of generated articles, leading to shallow, repetitive, and unoriginal outputs. To address these issues, we propose OmniThink, a machine writing framework that emulates the human-like process of iterative expansion and reflection. The core idea behind OmniThink is to simulate the cognitive behavior of learners as they progressively deepen their knowledge of the topics. Experimental results demonstrate that OmniThink improves the knowledge density of generated articles without compromising metrics such as coherence and depth. Human evaluations and expert feedback further highlight the potential of OmniThink to address real-world challenges in the generation of long-form articles.
☆ Enhancing Lexicon-Based Text Embeddings with Large Language Models
Recent large language models (LLMs) have demonstrated exceptional performance on general-purpose text embedding tasks. While dense embeddings have dominated related research, we introduce the first Lexicon-based EmbeddiNgS (LENS) leveraging LLMs that achieve competitive performance on these tasks. Regarding the inherent tokenization redundancy issue and unidirectional attention limitations in traditional causal LLMs, LENS consolidates the vocabulary space through token embedding clustering, and investigates bidirectional attention and various pooling strategies. Specifically, LENS simplifies lexicon matching by assigning each dimension to a specific token cluster, where semantically similar tokens are grouped together, and unlocking the full potential of LLMs through bidirectional attention. Extensive experiments demonstrate that LENS outperforms dense embeddings on the Massive Text Embedding Benchmark (MTEB), delivering compact feature representations that match the sizes of dense counterparts. Notably, combining LENSE with dense embeddings achieves state-of-the-art performance on the retrieval subset of MTEB (i.e. BEIR).
☆ Suggesting Code Edits in Interactive Machine Learning Notebooks Using Large Language Models
Machine learning developers frequently use interactive computational notebooks, such as Jupyter notebooks, to host code for data processing and model training. Jupyter notebooks provide a convenient tool for writing machine learning pipelines and interactively observing outputs, however, maintaining Jupyter notebooks, e.g., to add new features or fix bugs, can be challenging due to the length and complexity of the notebooks. Moreover, there is no existing benchmark related to developer edits on Jupyter notebooks. To address this, we present the first dataset of 48,398 Jupyter notebook edits derived from 20,095 revisions of 792 machine learning repositories on GitHub, and perform the first study of the using LLMs to predict code edits in Jupyter notebooks. Our dataset captures granular details of cell-level and line-level modifications, offering a foundation for understanding real-world maintenance patterns in machine learning workflows. We observed that the edits on Jupyter notebooks are highly localized, with changes averaging only 166 lines of code in repositories. While larger models outperform smaller counterparts in code editing, all models have low accuracy on our dataset even after finetuning, demonstrating the complexity of real-world machine learning maintenance tasks. Our findings emphasize the critical role of contextual information in improving model performance and point toward promising avenues for advancing large language models' capabilities in engineering machine learning code.
☆ Attention based Bidirectional GRU hybrid model for inappropriate content detection in Urdu language
With the increased use of the internet and social networks for online discussions, the spread of toxic and inappropriate content on social networking sites has also increased. Several studies have been conducted in different languages. However, there is less work done for South Asian languages for inappropriate content identification using deep learning techniques. In Urdu language, the spellings are not unique, and people write different common spellings for the same word, while mixing it other languages, like English in the text makes it more challenging, and limited research work is available to process such language with the finest algorithms. The use of attention layer with a deep learning model can help handling the long-term dependencies and increase its efficiency . To explore the effects of the attention layer, this study proposes attention-based Bidirectional GRU hybrid model for identifying inappropriate content in Urdu Unicode text language. Four different baseline deep learning models; LSTM, Bi-LSTM, GRU, and TCN, are used to compare the performance of the proposed model. The results of these models were compared based on evaluation metrics, dataset size, and impact of the word embedding layer. The pre-trained Urdu word2Vec embeddings were utilized for our case. Our proposed model BiGRU-A outperformed all other baseline models by yielding 84\% accuracy without using pre-trained word2Vec layer. From our experiments, we have established that the attention layer improves the model's efficiency, and pre-trained word2Vec embedding does not work well with an inappropriate content dataset.
☆ Comparative Insights from 12 Machine Learning Models in Extracting Economic Ideology from Political Text
This study conducts a systematic assessment of the capabilities of 12 machine learning models and model variations in detecting economic ideology. As an evaluation benchmark, I use manifesto data spanning six elections in the United Kingdom and pre-annotated by expert and crowd coders. The analysis assesses the performance of several generative, fine-tuned, and zero-shot models at the granular and aggregate levels. The results show that generative models such as GPT-4o and Gemini 1.5 Flash consistently outperform other models against all benchmarks. However, they pose issues of accessibility and resource availability. Fine-tuning yielded competitive performance and offers a reliable alternative through domain-specific optimization. But its dependency on training data severely limits scalability. Zero-shot models consistently face difficulties with identifying signals of economic ideology, often resulting in negative associations with human coding. Using general knowledge for the domain-specific task of ideology scaling proved to be unreliable. Other key findings include considerable within-party variation, fine-tuning benefiting from larger training data, and zero-shot's sensitivity to prompt content. The assessments include the strengths and limitations of each model and derive best-practices for automated analyses of political content.
☆ Domain Adaptation of Foundation LLMs for e-Commerce
We present the e-Llama models: 8 billion and 70 billion parameter large language models that are adapted towards the e-commerce domain. These models are meant as foundation models with deep knowledge about e-commerce, that form a base for instruction- and fine-tuning. The e-Llama models are obtained by continuously pretraining the Llama 3.1 base models on 1 trillion tokens of domain-specific data. We discuss our approach and motivate our choice of hyperparameters with a series of ablation studies. To quantify how well the models have been adapted to the e-commerce domain, we define and implement a set of multilingual, e-commerce specific evaluation tasks. We show that, when carefully choosing the training setup, the Llama 3.1 models can be adapted towards the new domain without sacrificing significant performance on general domain tasks. We also explore the possibility of merging the adapted model and the base model for a better control of the performance trade-off between domains.
☆ Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models
Language has long been conceived as an essential tool for human reasoning. The breakthrough of Large Language Models (LLMs) has sparked significant research interest in leveraging these models to tackle complex reasoning tasks. Researchers have moved beyond simple autoregressive token generation by introducing the concept of "thought" -- a sequence of tokens representing intermediate steps in the reasoning process. This innovative paradigm enables LLMs' to mimic complex human reasoning processes, such as tree search and reflective thinking. Recently, an emerging trend of learning to reason has applied reinforcement learning (RL) to train LLMs to master reasoning processes. This approach enables the automatic generation of high-quality reasoning trajectories through trial-and-error search algorithms, significantly expanding LLMs' reasoning capacity by providing substantially more training data. Furthermore, recent studies demonstrate that encouraging LLMs to "think" with more tokens during test-time inference can further significantly boost reasoning accuracy. Therefore, the train-time and test-time scaling combined to show a new research frontier -- a path toward Large Reasoning Model. The introduction of OpenAI's o1 series marks a significant milestone in this research direction. In this survey, we present a comprehensive review of recent progress in LLM reasoning. We begin by introducing the foundational background of LLMs and then explore the key technical components driving the development of large reasoning models, with a focus on automated data construction, learning-to-reason techniques, and test-time scaling. We also analyze popular open-source projects at building large reasoning models, and conclude with open challenges and future research directions.
comment: 36 pages, 5 figures
☆ The Heap: A Contamination-Free Multilingual Code Dataset for Evaluating Large Language Models
The recent rise in the popularity of large language models has spurred the development of extensive code datasets needed to train them. This has left limited code available for collection and use in the downstream investigation of specific behaviors, or evaluation of large language models without suffering from data contamination. To address this problem, we release The Heap, a large multilingual dataset covering 57 programming languages that has been deduplicated with respect to other open datasets of code, enabling researchers to conduct fair evaluations of large language models without significant data cleaning overhead.
comment: Pre-Print. Accepted to FORGE 2025 Dataset Track
☆ CarMem: Enhancing Long-Term Memory in LLM Voice Assistants through Category-Bounding COLING 2025
In today's assistant landscape, personalisation enhances interactions, fosters long-term relationships, and deepens engagement. However, many systems struggle with retaining user preferences, leading to repetitive user requests and disengagement. Furthermore, the unregulated and opaque extraction of user preferences in industry applications raises significant concerns about privacy and trust, especially in regions with stringent regulations like Europe. In response to these challenges, we propose a long-term memory system for voice assistants, structured around predefined categories. This approach leverages Large Language Models to efficiently extract, store, and retrieve preferences within these categories, ensuring both personalisation and transparency. We also introduce a synthetic multi-turn, multi-session conversation dataset (CarMem), grounded in real industry data, tailored to an in-car voice assistant setting. Benchmarked on the dataset, our system achieves an F1-score of .78 to .95 in preference extraction, depending on category granularity. Our maintenance strategy reduces redundant preferences by 95% and contradictory ones by 92%, while the accuracy of optimal retrieval is at .87. Collectively, the results demonstrate the system's suitability for industrial applications.
comment: Accepted for presentation at the International Conference on Computational Linguistics (COLING 2025)
☆ From Scarcity to Capability: Empowering Fake News Detection in Low-Resource Languages with LLMs
The rapid spread of fake news presents a significant global challenge, particularly in low-resource languages like Bangla, which lack adequate datasets and detection tools. Although manual fact-checking is accurate, it is expensive and slow to prevent the dissemination of fake news. Addressing this gap, we introduce BanFakeNews-2.0, a robust dataset to enhance Bangla fake news detection. This version includes 11,700 additional, meticulously curated fake news articles validated from credible sources, creating a proportional dataset of 47,000 authentic and 13,000 fake news items across 13 categories. In addition, we created a manually curated independent test set of 460 fake and 540 authentic news items for rigorous evaluation. We invest efforts in collecting fake news from credible sources and manually verified while preserving the linguistic richness. We develop a benchmark system utilizing transformer-based architectures, including fine-tuned Bidirectional Encoder Representations from Transformers variants (F1-87\%) and Large Language Models with Quantized Low-Rank Approximation (F1-89\%), that significantly outperforms traditional methods. BanFakeNews-2.0 offers a valuable resource to advance research and application in fake news detection for low-resourced languages. We publicly release our dataset and model on Github to foster research in this direction.
☆ Stylomech: Unveiling Authorship via Computational Stylometry in English and Romanized Sinhala
With the advent of Web 2.0, the development in social technology coupled with global communication systematically brought positive and negative impacts to society. Copyright claims and Author identification are deemed crucial as there has been a considerable amount of increase in content violation owing to the lack of proper ethics in society. The Author's attribution in both English and Romanized Sinhala became a major requirement in the last few decades. As an area largely unexplored, particularly within the context of Romanized Sinhala, the research contributes significantly to the field of computational linguistics. The proposed author attribution system offers a unique approach, allowing for the comparison of only two sets of text: suspect author and anonymous text, a departure from traditional methodologies which often rely on larger corpora. This work focuses on using the numerical representation of various pairs of the same and different authors allowing for, the model to train on these representations as opposed to text, this allows for it to apply to a multitude of authors and contexts, given that the suspected author text, and the anonymous text are of reasonable quality. By expanding the scope of authorship attribution to encompass diverse linguistic contexts, the work contributes to fostering trust and accountability in digital communication, especially in Sri Lanka. This research presents a pioneering approach to author attribution in both English and Romanized Sinhala, addressing a critical need for content verification and intellectual property rights enforcement in the digital age.
comment: 3 figure, 1 image
☆ Analyzing Continuous Semantic Shifts with Diachronic Word Similarity Matrices COLING2025
The meanings and relationships of words shift over time. This phenomenon is referred to as semantic shift.Research focused on understanding how semantic shifts occur over multiple time periods is essential for gaining a detailed understanding of semantic shifts.However, detecting change points only between adjacent time periods is insufficient for analyzing detailed semantic shifts, and using BERT-based methods to examine word sense proportions incurs a high computational cost.To address those issues, we propose a simple yet intuitive framework for how semantic shifts occur over multiple time periods by leveraging a similarity matrix between the embeddings of the same word through time.We compute a diachronic word similarity matrix using fast and lightweight word embeddings across arbitrary time periods, making it deeper to analyze continuous semantic shifts.Additionally, by clustering the similarity matrices for different words, we can categorize words that exhibit similar behavior of semantic shift in an unsupervised manner.
comment: COLING2025
☆ Confidence Estimation for Error Detection in Text-to-SQL Systems AAAI 2025
Text-to-SQL enables users to interact with databases through natural language, simplifying the retrieval and synthesis of information. Despite the success of large language models (LLMs) in converting natural language questions into SQL queries, their broader adoption is limited by two main challenges: achieving robust generalization across diverse queries and ensuring interpretative confidence in their predictions. To tackle these issues, our research investigates the integration of selective classifiers into Text-to-SQL systems. We analyse the trade-off between coverage and risk using entropy based confidence estimation with selective classifiers and assess its impact on the overall performance of Text-to-SQL models. Additionally, we explore the models' initial calibration and improve it with calibration techniques for better model alignment between confidence and accuracy. Our experimental results show that encoder-decoder T5 is better calibrated than in-context-learning GPT 4 and decoder-only Llama 3, thus the designated external entropy-based selective classifier has better performance. The study also reveal that, in terms of error detection, selective classifier with a higher probability detects errors associated with irrelevant questions rather than incorrect query generations.
comment: 15 pages, 11 figures, to be published in AAAI 2025 Proceedings
☆ Augmenting a Large Language Model with a Combination of Text and Visual Data for Conversational Visualization of Global Geospatial Data
We present a method for augmenting a Large Language Model (LLM) with a combination of text and visual data to enable accurate question answering in visualization of scientific data, making conversational visualization possible. LLMs struggle with tasks like visual data interaction, as they lack contextual visual information. We address this problem by merging a text description of a visualization and dataset with snapshots of the visualization. We extract their essential features into a structured text file, highly compact, yet descriptive enough to appropriately augment the LLM with contextual information, without any fine-tuning. This approach can be applied to any visualization that is already finally rendered, as long as it is associated with some textual description.
☆ PIER: A Novel Metric for Evaluating What Matters in Code-Switching ICASSP 2025
Code-switching, the alternation of languages within a single discourse, presents a significant challenge for Automatic Speech Recognition. Despite the unique nature of the task, performance is commonly measured with established metrics such as Word-Error-Rate (WER). However, in this paper, we question whether these general metrics accurately assess performance on code-switching. Specifically, using both Connectionist-Temporal-Classification and Encoder-Decoder models, we show fine-tuning on non-code-switched data from both matrix and embedded language improves classical metrics on code-switching test sets, although actual code-switched words worsen (as expected). Therefore, we propose Point-of-Interest Error Rate (PIER), a variant of WER that focuses only on specific words of interest. We instantiate PIER on code-switched utterances and show that this more accurately describes the code-switching performance, showing huge room for improvement in future work. This focused evaluation allows for a more precise assessment of model performance, particularly in challenging aspects such as inter-word and intra-word code-switching.
comment: Accepted at ICASSP 2025
☆ Exploring the Inquiry-Diagnosis Relationship with Advanced Patient Simulators
Online medical consultation (OMC) restricts doctors to gathering patient information solely through inquiries, making the already complex sequential decision-making process of diagnosis even more challenging. Recently, the rapid advancement of large language models has demonstrated a significant potential to transform OMC. However, most studies have primarily focused on improving diagnostic accuracy under conditions of relatively sufficient information, while paying limited attention to the "inquiry" phase of the consultation process. This lack of focus has left the relationship between "inquiry" and "diagnosis" insufficiently explored. In this paper, we first extract real patient interaction strategies from authentic doctor-patient conversations and use these strategies to guide the training of a patient simulator that closely mirrors real-world behavior. By inputting medical records into our patient simulator to simulate patient responses, we conduct extensive experiments to explore the relationship between "inquiry" and "diagnosis" in the consultation process. Experimental results demonstrate that inquiry and diagnosis adhere to the Liebig's law: poor inquiry quality limits the effectiveness of diagnosis, regardless of diagnostic capability, and vice versa. Furthermore, the experiments reveal significant differences in the inquiry performance of various models. To investigate this phenomenon, we categorize the inquiry process into four types: (1) chief complaint inquiry; (2) specification of known symptoms; (3) inquiry about accompanying symptoms; and (4) gathering family or medical history. We analyze the distribution of inquiries across the four types for different models to explore the reasons behind their significant performance differences. We plan to open-source the weights and related code of our patient simulator at https://github.com/LIO-H-ZEN/PatientSimulator.
☆ Scaling Graph-Based Dependency Parsing with Arc Vectorization and Attention-Based Refinement
We propose a novel architecture for graph-based dependency parsing that explicitly constructs vectors, from which both arcs and labels are scored. Our method addresses key limitations of the standard two-pipeline approach by unifying arc scoring and labeling into a single network, reducing scalability issues caused by the information bottleneck and lack of parameter sharing. Additionally, our architecture overcomes limited arc interactions with transformer layers to efficiently simulate higher-order dependencies. Experiments on PTB and UD show that our model outperforms state-of-the-art parsers in both accuracy and efficiency.
☆ Solving the unsolvable: Translating case law in Hong Kong
This paper addresses the challenges translating case law under Hong Kong's bilingual legal system. It highlights the initial success of translating all written statutes into Chinese before the 1997 handover, a task mandated by the Basic Law. The effort involved significant collaboration among legal, linguistic, and translation experts, resulting in a comprehensive and culturally appropriate bilingual legal system. However, translating case law remains a significant challenge due to the sheer volume and continuous growth of judicial decisions. The paper critiques the governments and judiciarys sporadic and uncoordinated efforts to translate case law, contrasting it with the thorough approach previously taken for statute translation. Although the government acknowledges the importance of legal bilingualism, it lacks a sustainable strategy for translating case law. The Judiciarys position that translating all judgments is unnecessary, unrealistic, and not cost-effectiveis analyzed and critiqued for its impact on legal transparency and public trust. A proposed solution involves leveraging machine translation technology through a human-machine interactive translation platform, which undergoes two major transitions. Initially based on a neural model, the platform transitions to using a large language model for improved translation accuracy. Furthermore, it evolves from a single-agent system to a multi-agent system, incorporating Translator, Annotator, and Proofreader agents. This multi-agent approach, supported by a grant, aims to facilitate efficient, high-quality translation of judicial judgments by integrating advanced artificial intelligence and continuous feedback mechanisms, thus better meeting the needs of a bilingual legal system.
☆ A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy
While large language models (LLMs) present significant potential for supporting numerous real-world applications and delivering positive social impacts, they still face significant challenges in terms of the inherent risk of privacy leakage, hallucinated outputs, and value misalignment, and can be maliciously used for generating toxic content and unethical purposes after been jailbroken. Therefore, in this survey, we present a comprehensive review of recent advancements aimed at mitigating these issues, organized across the four phases of LLM development and usage: data collecting and pre-training, fine-tuning and alignment, prompting and reasoning, and post-processing and auditing. We elaborate on the recent advances for enhancing the performance of LLMs in terms of privacy protection, hallucination reduction, value alignment, toxicity elimination, and jailbreak defenses. In contrast to previous surveys that focus on a single dimension of responsible LLMs, this survey presents a unified framework that encompasses these diverse dimensions, providing a comprehensive view of enhancing LLMs to better serve real-world applications.
☆ AutoCBT: An Autonomous Multi-agent Framework for Cognitive Behavioral Therapy in Psychological Counseling
Traditional in-person psychological counseling remains primarily niche, often chosen by individuals with psychological issues, while online automated counseling offers a potential solution for those hesitant to seek help due to feelings of shame. Cognitive Behavioral Therapy (CBT) is an essential and widely used approach in psychological counseling. The advent of large language models (LLMs) and agent technology enables automatic CBT diagnosis and treatment. However, current LLM-based CBT systems use agents with a fixed structure, limiting their self-optimization capabilities, or providing hollow, unhelpful suggestions due to redundant response patterns. In this work, we utilize Quora-like and YiXinLi single-round consultation models to build a general agent framework that generates high-quality responses for single-turn psychological consultation scenarios. We use a bilingual dataset to evaluate the quality of single-response consultations generated by each framework. Then, we incorporate dynamic routing and supervisory mechanisms inspired by real psychological counseling to construct a CBT-oriented autonomous multi-agent framework, demonstrating its general applicability. Experimental results indicate that AutoCBT can provide higher-quality automated psychological counseling services.
☆ Vision-Language Models Do Not Understand Negation
Many practical vision-language applications require models that understand negation, e.g., when using natural language to retrieve images which contain certain objects but not others. Despite advancements in vision-language models (VLMs) through large-scale training, their ability to comprehend negation remains underexplored. This study addresses the question: how well do current VLMs understand negation? We introduce NegBench, a new benchmark designed to evaluate negation understanding across 18 task variations and 79k examples spanning image, video, and medical datasets. The benchmark consists of two core tasks designed to evaluate negation understanding in diverse multimodal settings: Retrieval with Negation and Multiple Choice Questions with Negated Captions. Our evaluation reveals that modern VLMs struggle significantly with negation, often performing at chance level. To address these shortcomings, we explore a data-centric approach wherein we finetune CLIP models on large-scale synthetic datasets containing millions of negated captions. We show that this approach can result in a 10% increase in recall on negated queries and a 40% boost in accuracy on multiple-choice questions with negated captions.
comment: Project page: https://negbench.github.io
☆ mGeNTE: A Multilingual Resource for Gender-Neutral Language and Translation
Gender-neutral language reflects societal and linguistic shifts towards greater inclusivity by avoiding the implication that one gender is the norm over others. This is particularly relevant for grammatical gender languages, which heavily encode the gender of terms for human referents and over-relies on masculine forms, even when gender is unspecified or irrelevant. Language technologies are known to mirror these inequalities, being affected by a male bias and perpetuating stereotypical associations when translating into languages with extensive gendered morphology. In such cases, gender-neutral language can help avoid undue binary assumptions. However, despite its importance for creating fairer multi- and cross-lingual technologies, inclusive language research remains scarce and insufficiently supported in current resources. To address this gap, we present the multilingual mGeNTe dataset. Derived from the bilingual GeNTE (Piergentili et al., 2023), mGeNTE extends the original corpus to include the English-Italian/German/Spanish language pairs. Since each language pair is English-aligned with gendered and neutral sentences in the target languages, mGeNTE enables research in both automatic Gender-Neutral Translation (GNT) and language modelling for three grammatical gender languages.
☆ Evaluating LLM Abilities to Understand Tabular Electronic Health Records: A Comprehensive Study of Patient Data Extraction and Retrieval ECIR
Electronic Health Record (EHR) tables pose unique challenges among which is the presence of hidden contextual dependencies between medical features with a high level of data dimensionality and sparsity. This study presents the first investigation into the abilities of LLMs to comprehend EHRs for patient data extraction and retrieval. We conduct extensive experiments using the MIMICSQL dataset to explore the impact of the prompt structure, instruction, context, and demonstration, of two backbone LLMs, Llama2 and Meditron, based on task performance. Through quantitative and qualitative analyses, our findings show that optimal feature selection and serialization methods can enhance task performance by up to 26.79% compared to naive approaches. Similarly, in-context learning setups with relevant example selection improve data extraction performance by 5.95%. Based on our study findings, we propose guidelines that we believe would help the design of LLM-based models to support health search.
comment: To be published as full paper in the Proceedings of the European Conference on Information Retrieval (ECIR) 2025. Preprint
☆ ChartInsighter: An Approach for Mitigating Hallucination in Time-series Chart Summary Generation with A Benchmark Dataset
Effective chart summary can significantly reduce the time and effort decision makers spend interpreting charts, enabling precise and efficient communication of data insights. Previous studies have faced challenges in generating accurate and semantically rich summaries of time-series data charts. In this paper, we identify summary elements and common hallucination types in the generation of time-series chart summaries, which serve as our guidelines for automatic generation. We introduce ChartInsighter, which automatically generates chart summaries of time-series data, effectively reducing hallucinations in chart summary generation. Specifically, we assign multiple agents to generate the initial chart summary and collaborate iteratively, during which they invoke external data analysis modules to extract insights and compile them into a coherent summary. Additionally, we implement a self-consistency test method to validate and correct our summary. We create a high-quality benchmark of charts and summaries, with hallucination types annotated on a sentence-by-sentence basis, facilitating the evaluation of the effectiveness of reducing hallucinations. Our evaluations using our benchmark show that our method surpasses state-of-the-art models, and that our summary hallucination rate is the lowest, which effectively reduces various hallucinations and improves summary quality. The benchmark is available at https://github.com/wangfen01/ChartInsighter.
☆ Algorithm for Semantic Network Generation from Texts of Low Resource Languages Such as Kiswahili
Processing low-resource languages, such as Kiswahili, using machine learning is difficult due to lack of adequate training data. However, such low-resource languages are still important for human communication and are already in daily use and users need practical machine processing tasks such as summarization, disambiguation and even question answering (QA). One method of processing such languages, while bypassing the need for training data, is the use semantic networks. Some low resource languages, such as Kiswahili, are of the subject-verb-object (SVO) structure, and similarly semantic networks are a triple of subject-predicate-object, hence SVO parts of speech tags can map into a semantic network triple. An algorithm to process raw natural language text and map it into a semantic network is therefore necessary and desirable in structuring low resource languages texts. This algorithm tested on the Kiswahili QA task with upto 78.6% exact match.
comment: 18 pages, 3 figures, published in Open Journal for Information Technology
☆ Shape-Based Single Object Classification Using Ensemble Method Classifiers
Nowadays, more and more images are available. Annotation and retrieval of the images pose classification problems, where each class is defined as the group of database images labelled with a common semantic label. Various systems have been proposed for content-based retrieval, as well as for image classification and indexing. In this paper, a hierarchical classification framework has been proposed for bridging the semantic gap effectively and achieving multi-category image classification. A well known pre-processing and post-processing method was used and applied to three problems; image segmentation, object identification and image classification. The method was applied to classify single object images from Amazon and Google datasets. The classification was tested for four different classifiers; BayesNetwork (BN), Random Forest (RF), Bagging and Vote. The estimated classification accuracies ranged from 20% to 99% (using 10-fold cross validation). The Bagging classifier presents the best performance, followed by the Random Forest classifier.
☆ A Study of In-Context-Learning-Based Text-to-SQL Errors
Large language models (LLMs) have been adopted to perform text-to-SQL tasks, utilizing their in-context learning (ICL) capability to translate natural language questions into structured query language (SQL). However, such a technique faces correctness problems and requires efficient repairing solutions. In this paper, we conduct the first comprehensive study of text-to-SQL errors. Our study covers four representative ICL-based techniques, five basic repairing methods, two benchmarks, and two LLM settings. We find that text-to-SQL errors are widespread and summarize 29 error types of 7 categories. We also find that existing repairing attempts have limited correctness improvement at the cost of high computational overhead with many mis-repairs. Based on the findings, we propose MapleRepair, a novel text-to-SQL error detection and repairing framework. The evaluation demonstrates that MapleRepair outperforms existing solutions by repairing 13.8% more queries with neglectable mis-repairs and 67.4% less overhead.
☆ Understanding Mental Health Content on Social Media and Its Effect Towards Suicidal Ideation
This review underscores the critical need for effective strategies to identify and support individuals with suicidal ideation, exploiting technological innovations in ML and DL to further suicide prevention efforts. The study details the application of these technologies in analyzing vast amounts of unstructured social media data to detect linguistic patterns, keywords, phrases, tones, and contextual cues associated with suicidal thoughts. It explores various ML and DL models like SVMs, CNNs, LSTM, neural networks, and their effectiveness in interpreting complex data patterns and emotional nuances within text data. The review discusses the potential of these technologies to serve as a life-saving tool by identifying at-risk individuals through their digital traces. Furthermore, it evaluates the real-world effectiveness, limitations, and ethical considerations of employing these technologies for suicide prevention, stressing the importance of responsible development and usage. The study aims to fill critical knowledge gaps by analyzing recent studies, methodologies, tools, and techniques in this field. It highlights the importance of synthesizing current literature to inform practical tools and suicide prevention efforts, guiding innovation in reliable, ethical systems for early intervention. This research synthesis evaluates the intersection of technology and mental health, advocating for the ethical and responsible application of ML, DL, and NLP to offer life-saving potential worldwide while addressing challenges like generalizability, biases, privacy, and the need for further research to ensure these technologies do not exacerbate existing inequities and harms.
☆ Efficient Few-Shot Medical Image Analysis via Hierarchical Contrastive Vision-Language Learning
Few-shot learning in medical image classification presents a significant challenge due to the limited availability of annotated data and the complex nature of medical imagery. In this work, we propose Adaptive Vision-Language Fine-tuning with Hierarchical Contrastive Alignment (HiCA), a novel framework that leverages the capabilities of Large Vision-Language Models (LVLMs) for medical image analysis. HiCA introduces a two-stage fine-tuning strategy, combining domain-specific pretraining and hierarchical contrastive learning to align visual and textual representations at multiple levels. We evaluate our approach on two benchmark datasets, Chest X-ray and Breast Ultrasound, achieving state-of-the-art performance in both few-shot and zero-shot settings. Further analyses demonstrate the robustness, generalizability, and interpretability of our method, with substantial improvements in performance compared to existing baselines. Our work highlights the potential of hierarchical contrastive strategies in adapting LVLMs to the unique challenges of medical imaging tasks.
☆ To Retrieve or Not to Retrieve? Uncertainty Detection for Dynamic Retrieval Augmented Generation
Retrieval-Augmented Generation equips large language models with the capability to retrieve external knowledge, thereby mitigating hallucinations by incorporating information beyond the model's intrinsic abilities. However, most prior works have focused on invoking retrieval deterministically, which makes it unsuitable for tasks such as long-form question answering. Instead, dynamically performing retrieval by invoking it only when the underlying LLM lacks the required knowledge can be more efficient. In this context, we delve deeper into the question, "To Retrieve or Not to Retrieve?" by exploring multiple uncertainty detection methods. We evaluate these methods for the task of long-form question answering, employing dynamic retrieval, and present our comparisons. Our findings suggest that uncertainty detection metrics, such as Degree Matrix Jaccard and Eccentricity, can reduce the number of retrieval calls by almost half, with only a slight reduction in question-answering accuracy.
☆ Perspective Transition of Large Language Models for Solving Subjective Tasks
Large language models (LLMs) have revolutionized the field of natural language processing, enabling remarkable progress in various tasks. Different from objective tasks such as commonsense reasoning and arithmetic question-answering, the performance of LLMs on subjective tasks is still limited, where the perspective on the specific problem plays crucial roles for better interpreting the context and giving proper response. For example, in certain scenarios, LLMs may perform better when answering from an expert role perspective, potentially eliciting their relevant domain knowledge. In contrast, in some scenarios, LLMs may provide more accurate responses when answering from a third-person standpoint, enabling a more comprehensive understanding of the problem and potentially mitigating inherent biases. In this paper, we propose Reasoning through Perspective Transition (RPT), a method based on in-context learning that enables LLMs to dynamically select among direct, role, and third-person perspectives for the best way to solve corresponding subjective problem. Through extensive experiments on totally 12 subjective tasks by using both closed-source and open-source LLMs including GPT-4, GPT-3.5, Llama-3, and Qwen-2, our method outperforms widely used single fixed perspective based methods such as chain-of-thought prompting and expert prompting, highlights the intricate ways that LLMs can adapt their perspectives to provide nuanced and contextually appropriate responses for different problems.
☆ Delayed Fusion: Integrating Large Language Models into First-Pass Decoding in End-to-end Speech Recognition ICASSP2025
This paper presents an efficient decoding approach for end-to-end automatic speech recognition (E2E-ASR) with large language models (LLMs). Although shallow fusion is the most common approach to incorporate language models into E2E-ASR decoding, we face two practical problems with LLMs. (1) LLM inference is computationally costly. (2) There may be a vocabulary mismatch between the ASR model and the LLM. To resolve this mismatch, we need to retrain the ASR model and/or the LLM, which is at best time-consuming and in many cases not feasible. We propose "delayed fusion," which applies LLM scores to ASR hypotheses with a delay during decoding and enables easier use of pre-trained LLMs in ASR tasks. This method can reduce not only the number of hypotheses scored by the LLM but also the number of LLM inference calls. It also allows re-tokenizion of ASR hypotheses during decoding if ASR and LLM employ different tokenizations. We demonstrate that delayed fusion provides improved decoding speed and accuracy compared to shallow fusion and N-best rescoring using the LibriHeavy ASR corpus and three public LLMs, OpenLLaMA 3B & 7B and Mistral 7B.
comment: Accepted to ICASSP2025
☆ Foundations of Large Language Models
This is a book about large language models. As indicated by the title, it primarily focuses on foundational concepts rather than comprehensive coverage of all cutting-edge technologies. The book is structured into four main chapters, each exploring a key area: pre-training, generative models, prompting techniques, and alignment methods. It is intended for college students, professionals, and practitioners in natural language processing and related fields, and can serve as a reference for anyone interested in large language models.
☆ A Simple Graph Contrastive Learning Framework for Short Text Classification AAAI2025
Short text classification has gained significant attention in the information age due to its prevalence and real-world applications. Recent advancements in graph learning combined with contrastive learning have shown promising results in addressing the challenges of semantic sparsity and limited labeled data in short text classification. However, existing models have certain limitations. They rely on explicit data augmentation techniques to generate contrastive views, resulting in semantic corruption and noise. Additionally, these models only focus on learning the intrinsic consistency between the generated views, neglecting valuable discriminative information from other potential views. To address these issues, we propose a Simple graph contrastive learning framework for Short Text Classification (SimSTC). Our approach involves performing graph learning on multiple text-related component graphs to obtain multi-view text embeddings. Subsequently, we directly apply contrastive learning on these embeddings. Notably, our method eliminates the need for data augmentation operations to generate contrastive views while still leveraging the benefits of multi-view contrastive learning. Despite its simplicity, our model achieves outstanding performance, surpassing large language models on various datasets.
comment: AAAI2025
☆ Boosting Short Text Classification with Multi-Source Information Exploration and Dual-Level Contrastive Learning AAAI2025
Short text classification, as a research subtopic in natural language processing, is more challenging due to its semantic sparsity and insufficient labeled samples in practical scenarios. We propose a novel model named MI-DELIGHT for short text classification in this work. Specifically, it first performs multi-source information (i.e., statistical information, linguistic information, and factual information) exploration to alleviate the sparsity issues. Then, the graph learning approach is adopted to learn the representation of short texts, which are presented in graph forms. Moreover, we introduce a dual-level (i.e., instance-level and cluster-level) contrastive learning auxiliary task to effectively capture different-grained contrastive information within massive unlabeled data. Meanwhile, previous models merely perform the main task and auxiliary tasks in parallel, without considering the relationship among tasks. Therefore, we introduce a hierarchical architecture to explicitly model the correlations between tasks. We conduct extensive experiments across various benchmark datasets, demonstrating that MI-DELIGHT significantly surpasses previous competitive models. It even outperforms popular large language models on several datasets.
comment: AAAI2025
☆ FineMedLM-o1: Enhancing the Medical Reasoning Ability of LLM from Supervised Fine-Tuning to Test-Time Training
Recent advancements in large language models (LLMs) have shown promise in medical applications such as disease diagnosis and treatment planning. However, most existing medical LLMs struggle with the advanced reasoning required for complex clinical scenarios, such as differential diagnosis or personalized treatment suggestions. We proposed FineMedLM-o1, which leverages high-quality synthetic medical data and long-form reasoning data for Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO), enabling advanced dialogue and deep reasoning capabilities. Additionally, we introduced Test-Time Training (TTT) in the medical domain for the first time, facilitating domain adaptation and ensuring reliable, accurate reasoning. Experimental results demonstrate that FineMedLM-o1 achieves a 23% average performance improvement over prior models on key medical benchmarks. Furthermore, the introduction of TTT provides an additional 14% performance boost, highlighting its effectiveness in enhancing medical reasoning capabilities. To support this process, we also proposed a novel method for synthesizing medical dialogue. Compared to other open-source datasets, our dataset stands out as superior in both quality and complexity. The project and data will be released on GitHub.
☆ Bridging Language Barriers in Healthcare: A Study on Arabic LLMs
This paper investigates the challenges of developing large language models (LLMs) proficient in both multilingual understanding and medical knowledge. We demonstrate that simply translating medical data does not guarantee strong performance on clinical tasks in the target language. Our experiments reveal that the optimal language mix in training data varies significantly across different medical tasks. We find that larger models with carefully calibrated language ratios achieve superior performance on native-language clinical tasks. Furthermore, our results suggest that relying solely on fine-tuning may not be the most effective approach for incorporating new language knowledge into LLMs. Instead, data and computationally intensive pretraining methods may still be necessary to achieve optimal performance in multilingual medical settings. These findings provide valuable guidance for building effective and inclusive medical AI systems for diverse linguistic communities.
☆ Qwen it detect machine-generated text?
This paper describes the approach of the Unibuc - NLP team in tackling the Coling 2025 GenAI Workshop, Task 1: Binary Multilingual Machine-Generated Text Detection. We explored both masked language models and causal models. For Subtask A, our best model achieved first-place out of 36 teams when looking at F1 Micro (Auxiliary Score) of 0.8333, and second-place when looking at F1 Macro (Main Score) of 0.8301
☆ Enhancing Generalization in Chain of Thought Reasoning for Smaller Models
Chain-of-Thought (CoT) reasoning in smaller language models is a challenging natural language process problem yet highly desirable in many real-life applications. Existing CoT knowledge distillation methods often suffer from overly conservative memorization in smaller LLMs, leading to low generalization confidence. As fully preserving the CoT ability of teacher model is impossible, we hypothesize that adversarial CoT fine-tuning is crucial for developing smaller LLM with robust CoT generalization. To this end, we propose \textit{PRompt-Assisted Domain-Adversarial fine-tuning} (PRADA), a principled fine-tuning framework that integrates diverse CoT domains. Specifically, PRADA pioneers two CoT improvements in smaller LLM: (1) Recovering the domain-invariant feature insight which typically lost during distillation with domain adversarial fine-tuning; (2) Enhancing the domain adaptability of CoT prompt engineering by employing domain-adversarial approaches. We theoretically demonstrate the effectiveness of our approach and empirically show that it significantly outperforms the state of the arts in a wide range of tasks. Moreover, our empirical findings reveal that the smaller LLM, when leveraging PRADA, aligns closely with domain knowledge, thereby improving the explainability of our approach.
☆ Conversational Text Extraction with Large Language Models Using Retrieval-Augmented Systems
This study introduces a system leveraging Large Language Models (LLMs) to extract text and enhance user interaction with PDF documents via a conversational interface. Utilizing Retrieval-Augmented Generation (RAG), the system provides informative responses to user inquiries while highlighting relevant passages within the PDF. Upon user upload, the system processes the PDF, employing sentence embeddings to create a document-specific vector store. This vector store enables efficient retrieval of pertinent sections in response to user queries. The LLM then engages in a conversational exchange, using the retrieved information to extract text and generate comprehensive, contextually aware answers. While our approach demonstrates competitive ROUGE values compared to existing state-of-the-art techniques for text extraction and summarization, we acknowledge that further qualitative evaluation is necessary to fully assess its effectiveness in real-world applications. The proposed system gives competitive ROUGE values as compared to existing state-of-the-art techniques for text extraction and summarization, thus offering a valuable tool for researchers, students, and anyone seeking to efficiently extract knowledge and gain insights from documents through an intuitive question-answering interface.
☆ Computing Optimization-Based Prompt Injections Against Closed-Weights Models By Misusing a Fine-Tuning API
We surface a new threat to closed-weight Large Language Models (LLMs) that enables an attacker to compute optimization-based prompt injections. Specifically, we characterize how an attacker can leverage the loss-like information returned from the remote fine-tuning interface to guide the search for adversarial prompts. The fine-tuning interface is hosted by an LLM vendor and allows developers to fine-tune LLMs for their tasks, thus providing utility, but also exposes enough information for an attacker to compute adversarial prompts. Through an experimental analysis, we characterize the loss-like values returned by the Gemini fine-tuning API and demonstrate that they provide a useful signal for discrete optimization of adversarial prompts using a greedy search algorithm. Using the PurpleLlama prompt injection benchmark, we demonstrate attack success rates between 65% and 82% on Google's Gemini family of LLMs. These attacks exploit the classic utility-security tradeoff - the fine-tuning interface provides a useful feature for developers but also exposes the LLMs to powerful attacks.
☆ Sentiment Analysis in Twitter Social Network Centered on Cryptocurrencies Using Machine Learning
Cryptocurrency is a digital currency that uses blockchain technology with secure encryption. Due to the decentralization of these currencies, traditional monetary systems and the capital market of each they, can influence a society. Therefore, due to the importance of the issue, the need to understand public opinion and analyze people's opinions in this regard increases. To understand the opinions and views of people about different topics, you can take help from social networks because they are a rich source of opinions. The Twitter social network is one of the main platforms where users discuss various topics, therefore, in the shortest time and with the lowest cost, the opinion of the community can be measured on this social network. Twitter Sentiment Analysis (TSA) is a field that analyzes the sentiment expressed in tweets. Considering that most of TSA's research efforts on cryptocurrencies are focused on English language, the purpose of this paper is to investigate the opinions of Iranian users on the Twitter social network about cryptocurrencies and provide the best model for classifying tweets based on sentiment. In the case of automatic analysis of tweets, managers and officials in the field of economy can gain knowledge from the general public's point of view about this issue and use the information obtained in order to properly manage this phenomenon. For this purpose, in this paper, in order to build emotion classification models, natural language processing techniques such as bag of words (BOW) and FastText for text vectorization and classical machine learning algorithms including KNN, SVM and Adaboost learning methods Deep including LSTM and BERT model were used for classification, and finally BERT linguistic model had the best accuracy with 83.50%.
comment: 6 pages and 5 figures
☆ Multiple Choice Questions: Reasoning Makes Large Language Models (LLMs) More Self-Confident Even When They Are Wrong
One of the most widely used methods to evaluate LLMs are Multiple Choice Question (MCQ) tests. MCQ benchmarks enable the testing of LLM knowledge on almost any topic at scale as the results can be processed automatically. To help the LLM answer, a few examples called few shots can be included in the prompt. Moreover, the LLM can be asked to answer the question directly with the selected option or to first provide the reasoning and then the selected answer, which is known as chain of thought. In addition to checking whether the selected answer is correct, the evaluation can look at the LLM-estimated probability of its response as an indication of the confidence of the LLM in the response. In this paper, we study how the LLM confidence in its answer depends on whether the model has been asked to answer directly or to provide the reasoning before answering. The results of the evaluation of questions on a wide range of topics in seven different models show that LLMs are more confident in their answers when they provide reasoning before the answer. This occurs regardless of whether the selected answer is correct. Our hypothesis is that this behavior is due to the reasoning that modifies the probability of the selected answer, as the LLM predicts the answer based on the input question and the reasoning that supports the selection made. Therefore, LLM estimated probabilities seem to have intrinsic limitations that should be understood in order to use them in evaluation procedures. Interestingly, the same behavior has been observed in humans, for whom explaining an answer increases confidence in its correctness.
♻ ☆ NL2KQL: From Natural Language to Kusto Query
Data is growing rapidly in volume and complexity. Proficiency in database query languages is pivotal for crafting effective queries. As coding assistants become more prevalent, there is significant opportunity to enhance database query languages. The Kusto Query Language (KQL) is a widely used query language for large semi-structured data such as logs, telemetries, and time-series for big data analytics platforms. This paper introduces NL2KQL an innovative framework that uses large language models (LLMs) to convert natural language queries (NLQs) to KQL queries. The proposed NL2KQL framework includes several key components: Schema Refiner which narrows down the schema to its most pertinent elements; the Few-shot Selector which dynamically selects relevant examples from a few-shot dataset; and the Query Refiner which repairs syntactic and semantic errors in KQL queries. Additionally, this study outlines a method for generating large datasets of synthetic NLQ-KQL pairs which are valid within a specific database contexts. To validate NL2KQL's performance, we utilize an array of online (based on query execution) and offline (based on query parsing) metrics. Through ablation studies, the significance of each framework component is examined, and the datasets used for benchmarking are made publicly available. This work is the first of its kind and is compared with available baselines to demonstrate its effectiveness.
♻ ☆ Aligning Brain Activity with Advanced Transformer Models: Exploring the Role of Punctuation in Semantic Processing
This research examines the congruence between neural activity and advanced transformer models, emphasizing the semantic significance of punctuation in text understanding. Utilizing an innovative approach originally proposed by Toneva and Wehbe, we evaluate four advanced transformer models RoBERTa, DistiliBERT, ALBERT, and ELECTRA against neural activity data. Our findings indicate that RoBERTa exhibits the closest alignment with neural activity, surpassing BERT in accuracy. Furthermore, we investigate the impact of punctuation removal on model performance and neural alignment, revealing that BERT's accuracy enhances in the absence of punctuation. This study contributes to the comprehension of how neural networks represent language and the influence of punctuation on semantic processing within the human brain.
♻ ☆ ReFactor GNNs: Revisiting Factorisation-based Models from a Message-Passing Perspective NeurIPS 2022
Factorisation-based Models (FMs), such as DistMult, have enjoyed enduring success for Knowledge Graph Completion (KGC) tasks, often outperforming Graph Neural Networks (GNNs). However, unlike GNNs, FMs struggle to incorporate node features and generalise to unseen nodes in inductive settings. Our work bridges the gap between FMs and GNNs by proposing ReFactor GNNs. This new architecture draws upon both modelling paradigms, which previously were largely thought of as disjoint. Concretely, using a message-passing formalism, we show how FMs can be cast as GNNs by reformulating the gradient descent procedure as message-passing operations, which forms the basis of our ReFactor GNNs. Across a multitude of well-established KGC benchmarks, our ReFactor GNNs achieve comparable transductive performance to FMs, and state-of-the-art inductive performance while using an order of magnitude fewer parameters.
comment: 36th Conference on Neural Information Processing Systems (NeurIPS 2022)
♻ ☆ PolInterviews -- A Dataset of German Politician Public Broadcast Interviews
This paper presents a novel dataset of public broadcast interviews featuring high-ranking German politicians. The interviews were sourced from YouTube, transcribed, processed for speaker identification, and stored in a tidy and open format. The dataset comprises 99 interviews with 33 different German politicians across five major interview formats, containing a total of 28,146 sentences. As the first of its kind, this dataset offers valuable opportunities for research on various aspects of political communication in the (German) political contexts, such as agenda-setting, interviewer dynamics, or politicians' self-presentation.
♻ ☆ Crafting Customisable Characters with LLMs: Introducing SimsChat, a Persona-Driven Role-Playing Agent Framework
Large Language Models (LLMs) demonstrate remarkable ability to comprehend instructions and generate human-like text, enabling sophisticated agent simulation beyond basic behavior replication. However, the potential for creating freely customisable characters remains underexplored. We introduce the Customisable Conversation Agent Framework, which employs LLMs to simulate real-world characters through personalised characteristic feature injection, enabling diverse character creation according to user preferences. We propose the SimsConv dataset, comprising 68 customised characters and 13,971 multi-turn role-playing dialogues across 1,360 real-world scenes. Characters are initially customised using pre-defined elements (career, aspiration, traits, skills), then expanded through personal and social profiles. Building on this, we present SimsChat, a freely customisable role-playing agent incorporating various realistic settings and topic-specified character interactions. Experimental results on both SimsConv and WikiRoleEval datasets demonstrate SimsChat's superior performance in maintaining character consistency, knowledge accuracy, and appropriate question rejection compared to existing models. Our framework provides valuable insights for developing more accurate and customisable human simulacra. Our data and code are publicly available at https://github.com/Bernard-Yang/SimsChat.
♻ ☆ Can linguists better understand DNA?
Multilingual transfer ability, which reflects how well models fine-tuned on one source language can be applied to other languages, has been well studied in multilingual pre-trained models. However, the existence of such capability transfer between natural language and gene sequences/languages remains under explored.This study addresses this gap by drawing inspiration from the sentence-pair classification task used for evaluating sentence similarity in natural language. We constructed two analogous tasks: DNA-pair classification(DNA sequence similarity) and DNA-protein-pair classification(gene coding determination). These tasks were designed to validate the transferability of capabilities from natural language to gene sequences. Even a small-scale pre-trained model like GPT-2-small, which was pre-trained on English, achieved an accuracy of 78% on the DNA-pair classification task after being fine-tuned on English sentence-pair classification data(XTREME PAWS-X). While training a BERT model on multilingual text, the precision reached 89%. On the more complex DNA-protein-pair classification task, however, the model's output was barely distinguishable from random output.Experimental validation has confirmed that the transfer of capabilities from natural language to biological language is unequivocally present. Building on this foundation, we have also investigated the impact of model parameter scale and pre-training on this capability transfer. We provide recommendations for facilitating the transfer of capabilities from natural language to genetic language,as well as new approaches for conducting biological research based on this capability.This study offers an intriguing new perspective on exploring the relationship between natural language and genetic language.
comment: 20 pages,7 figures
♻ ☆ aiXcoder-7B: A Lightweight and Effective Large Language Model for Code Processing ICSE 2025
Large Language Models (LLMs) have been widely used in code completion, and researchers are focusing on scaling up LLMs to improve their accuracy. However, larger LLMs have lower inference efficiency, affecting developers' experience and productivity. In this paper, we propose a lightweight and effective LLM for code completion named aiXcoder-7B. Compared to existing LLMs, aiXcoder-7B achieves higher code completion accuracy while having smaller scales (i.e., 7 billion parameters). We attribute the superiority of aiXcoder-7B to three key factors: (1) Multi-objective training. We employ three training objectives, one of which is our proposed Structured Fill-In-the-Middle (SFIM). SFIM considers the syntax structures in code and effectively improves the performance of LLMs for code. (2) Diverse data sampling strategies. They consider inter-file relationships and enhance the capability of LLMs in understanding cross-file contexts. (3) Extensive high-quality data. We establish a rigorous data collection pipeline and consume a total of 1.2 trillion unique tokens for training aiXcoder-7B. This vast volume of data enables aiXcoder-7B to learn a broad distribution of code. We evaluate aiXcoder-7B in five popular code completion benchmarks and a new benchmark collected by this paper. The results show that aiXcoder-7B outperforms the latest six LLMs with similar sizes and even surpasses four larger LLMs (e.g., StarCoder2-15B and CodeLlama-34B), positioning aiXcoder-7B as a lightweight and effective LLM for academia and industry. Finally, we summarize three valuable insights for helping practitioners train the next generations of LLMs for code. aiXcoder-7B has been open-souced and gained significant attention. Until January 2025, aiXcoder-7B has received 2,226 GitHub Stars.
comment: (1) Accepted by the 47th International Conference on Software Engineering (ICSE 2025). (2) aiXcoder-7B is available at https://github.com/aixcoder-plugin/aiXcoder-7B
♻ ☆ AudioBERT: Audio Knowledge Augmented Language Model ICASSP 2025
Recent studies have identified that language models, pretrained on text-only datasets, often lack elementary visual knowledge, \textit{e.g.,} colors of everyday objects. Motivated by this observation, we ask whether a similar shortcoming exists in terms of the \textit{auditory} knowledge. To answer this question, we construct a new dataset called AuditoryBench, which consists of two novel tasks for evaluating auditory knowledge. Based on our analysis using the benchmark, we find that language models also suffer from a severe lack of auditory knowledge. To address this limitation, we propose AudioBERT, a novel method to augment the auditory knowledge of BERT through a retrieval-based approach. First, we detect auditory knowledge spans in prompts to query our retrieval model efficiently. Then, we inject audio knowledge into BERT and switch on low-rank adaptation for effective adaptation when audio knowledge is required. Our experiments demonstrate that AudioBERT is quite effective, achieving superior performance on the AuditoryBench. The dataset and code are available at \bulurl{https://github.com/HJ-Ok/AudioBERT}.
comment: 5 pages, 3 figures, ICASSP 2025
♻ ☆ Focus On This, Not That! Steering LLMs With Adaptive Feature Specification
Despite the success of Instruction Tuning (IT) in training large language models (LLMs) to perform arbitrary user-specified tasks, these models often still leverage spurious or biased features learned from their training data, leading to undesired behaviours when deploying them in new contexts. In this work, we introduce Focus Instruction Tuning (FIT), which trains LLMs to condition their responses by focusing on specific features whilst ignoring others, leading to different behaviours based on what features are specified. Across several experimental settings, we show that focus-tuned models can be adaptively steered by focusing on different features at inference-time: for instance, robustness can be improved by focusing on task-causal features and ignoring spurious features, and social bias can be mitigated by ignoring demographic categories. Furthermore, FIT can steer behaviour in new contexts, generalising under distribution shift and to new unseen features at inference time, and thereby facilitating more robust, fair, and controllable LLM applications in real-world environments.
comment: 28pages, 14 figures
♻ ☆ Leveraging Fine-Tuned Retrieval-Augmented Generation with Long-Context Support: For 3GPP Standards
Recent studies show that large language models (LLMs) struggle with technical standards in telecommunications. We propose a fine-tuned retrieval-augmented generation (RAG) system based on the Phi-2 small language model (SLM) to serve as an oracle for communication networks. Our developed system leverages forward-looking semantic chunking to adaptively determine parsing breakpoints based on embedding similarity, enabling effective processing of diverse document formats. To handle the challenge of multiple similar contexts in technical standards, we employ a re-ranking algorithm to prioritize the most relevant retrieved chunks. Recognizing the limitations of Phi-2's small context window, we implement a recent technique, namely SelfExtend, to expand the context window during inference, which not only boosts the performance but also can accommodate a wider range of user queries and design requirements from customers to specialized technicians. For fine-tuning, we utilize the low-rank adaptation (LoRA) technique to enhance computational efficiency during training and enable effective fine-tuning on small datasets. Our comprehensive experiments demonstrate substantial improvements over existing question-answering approaches in the telecom domain, achieving performance that exceeds larger language models such as GPT-4 (which is about 880 times larger in size). This work presents a novel approach to leveraging SLMs for communication networks, offering a balance of efficiency and performance. This work can serve as a foundation towards agentic language models for networks.
comment: submitted to Proc. IEEE Globecom
♻ ☆ RAGBench: Explainable Benchmark for Retrieval-Augmented Generation Systems
Retrieval-Augmented Generation (RAG) has become a standard architectural pattern for incorporating domain-specific knowledge into user-facing chat applications powered by Large Language Models (LLMs). RAG systems are characterized by (1) a document retriever that queries a domain-specific corpus for context information relevant to an input query, and (2) an LLM that generates a response based on the provided query and context. However, comprehensive evaluation of RAG systems remains a challenge due to the lack of unified evaluation criteria and annotated datasets. In response, we introduce RAGBench: the first comprehensive, large-scale RAG benchmark dataset of 100k examples. It covers five unique industry-specific domains and various RAG task types. RAGBench examples are sourced from industry corpora such as user manuals, making it particularly relevant for industry applications. Further, we formalize the TRACe evaluation framework: a set of explainable and actionable RAG evaluation metrics applicable across all RAG domains. We release the labeled dataset at https://huggingface.co/datasets/rungalileo/ragbench. RAGBench explainable labels facilitate holistic evaluation of RAG systems, enabling actionable feedback for continuous improvement of production applications. Thorough extensive benchmarking, we find that LLM-based RAG evaluation methods struggle to compete with a finetuned RoBERTa model on the RAG evaluation task. We identify areas where existing approaches fall short and propose the adoption of RAGBench with TRACe towards advancing the state of RAG evaluation systems.
♻ ☆ SD-Eval: A Benchmark Dataset for Spoken Dialogue Understanding Beyond Words NeurIPS 2024
Speech encompasses a wealth of information, including but not limited to content, paralinguistic, and environmental information. This comprehensive nature of speech significantly impacts communication and is crucial for human-computer interaction. Chat-Oriented Large Language Models (LLMs), known for their general-purpose assistance capabilities, have evolved to handle multi-modal inputs, including speech. Although these models can be adept at recognizing and analyzing speech, they often fall short of generating appropriate responses. We argue that this is due to the lack of principles on task definition and model development, which requires open-source datasets and metrics suitable for model evaluation. To bridge the gap, we present SD-Eval, a benchmark dataset aimed at multidimensional evaluation of spoken dialogue understanding and generation. SD-Eval focuses on paralinguistic and environmental information and includes 7,303 utterances, amounting to 8.76 hours of speech data. The data is aggregated from eight public datasets, representing four perspectives: emotion, accent, age, and background sound. To assess the SD-Eval benchmark dataset, we implement three different models and construct a training set following a process similar to that of SD-Eval. The training set contains 1,052.72 hours of speech data and 724.4k utterances. We also conduct a comprehensive evaluation using objective evaluation methods (e.g. BLEU and ROUGE), subjective evaluations and LLM-based metrics for the generated responses. Models conditioned with paralinguistic and environmental information outperform their counterparts in both objective and subjective measures. Moreover, experiments demonstrate that LLM-based metrics show a higher correlation with human evaluation compared to traditional metrics. We open-source SD-Eval at https://github.com/amphionspace/SD-Eval.
comment: Accepted to NeurIPS 2024
♻ ☆ Discriminative Representation learning via Attention-Enhanced Contrastive Learning for Short Text Clustering
Contrastive learning has gained significant attention in short text clustering, yet it has an inherent drawback of mistakenly identifying samples from the same category as negatives and then separating them in the feature space (false negative separation), which hinders the generation of superior representations. To generate more discriminative representations for efficient clustering, we propose a novel short text clustering method, called Discriminative Representation learning via \textbf{A}ttention-\textbf{E}nhanced \textbf{C}ontrastive \textbf{L}earning for Short Text Clustering (\textbf{AECL}). The \textbf{AECL} consists of two modules which are the pseudo-label generation module and the contrastive learning module. Both modules build a sample-level attention mechanism to capture similarity relationships between samples and aggregate cross-sample features to generate consistent representations. Then, the former module uses the more discriminative consistent representation to produce reliable supervision information for assist clustering, while the latter module explores similarity relationships and consistent representations optimize the construction of positive samples to perform similarity-guided contrastive learning, effectively addressing the false negative separation issue. Experimental results demonstrate that the proposed \textbf{AECL} outperforms state-of-the-art methods. If the paper is accepted, we will open-source the code.
♻ ☆ Collaborative Gym: A Framework for Enabling and Evaluating Human-Agent Collaboration
Recent advancements in language models (LMs) have sparked growing interest in developing LM agents. While fully autonomous agents could excel in many scenarios, numerous use cases inherently require them to collaborate with humans due to humans' latent preferences, domain expertise, or need for control. To facilitate the study of human-agent collaboration, we present Collaborative Gym (Co-Gym), a general framework enabling asynchronous, tripartite interaction among agents, humans, and task environments. We instantiate Co-Gym with three representative tasks in both simulated and real-world conditions, and propose an evaluation framework that assesses both the collaboration outcomes and processes. Our findings reveal that collaborative agents consistently outperform their fully autonomous counterparts in task performance within those delivered cases, achieving win rates of 86% in Travel Planning, 74% in Tabular Analysis, and 66% in Related Work when evaluated by real users. However, our study also highlights significant challenges in developing collaborative agents, requiring advancements in core aspects of intelligence -- communication capabilities, situational awareness, and balancing autonomy and human control.
comment: Preprint. Work in progress
♻ ☆ MERaLiON-TextLLM: Cross-Lingual Understanding of Large Language Models in Chinese, Indonesian, Malay, and Singlish
Multilingual large language models (MLLMs) have shown impressive capabilities across a variety of languages. However, efficacy can differ greatly between different language families, especially for those with limited linguistic resources. This report presents MERaLiON-TextLLM, a series of open-source language models specifically tailored to improve understanding and generation in Chinese, Indonesian, Malay, and Singlish. The initial released model is built on Llama-3-8B-Base and refined through a meticulously crafted process of continued pre-training and weight merging. Our approach achieves performance improvements across benchmarks in these languages, exceeding the capabilities of the official Llama-3 models. We provide the model checkpoints as a resource to support further research and development in cross-lingual language understanding.
♻ ☆ Do LLMs Really Think Step-by-step In Implicit Reasoning?
It has been well-known that Chain-of-Thought can remarkably enhance LLMs' performance on complex tasks. However, because it also introduces slower inference speeds and higher computational costs, many researches have attempted to use implicit CoT, which does not need LLMs to explicitly generate the intermediate steps. However, the invisible reasoning process leaves us a doubt that, can implicit CoT really be equal to explicit CoT? Therefore, in this study, we address this question through experiments. We probe the information of intermediate steps from the model's hidden states when it is either trained or prompted to perform implicit CoT. The results surprisingly indicate that when prompted, LLMs hardly think about intermediate steps, suggesting they may just rely on experience rather than strict step-by-step reasoning. But when trained, they indeed calculate intermediate steps. Moreover, in both situations, we find the effect of using implicit CoT is susceptible to the format of the problem, reaffirming the current deficiency of implicit CoT.
comment: The code is in https://github.com/yuyijiong/if_step_by_step_implicit_CoT
♻ ☆ MERaLiON-AudioLLM: Bridging Audio and Language with Large Language Models
We introduce MERaLiON-AudioLLM (Multimodal Empathetic Reasoning and Learning in One Network), the first speech-text model tailored for Singapore's multilingual and multicultural landscape. Developed under the National Large Language Models Funding Initiative, Singapore, MERaLiON-AudioLLM integrates advanced speech and text processing to address the diverse linguistic nuances of local accents and dialects, enhancing accessibility and usability in complex, multilingual environments. Our results demonstrate improvements in both speech recognition and task-specific understanding, positioning MERaLiON-AudioLLM as a pioneering solution for region specific AI applications. We envision this release to set a precedent for future models designed to address localised linguistic and cultural contexts in a global framework.
comment: https://huggingface.co/MERaLiON/MERaLiON-AudioLLM-Whisper-SEA-LION
♻ ☆ CrisisSense-LLM: Instruction Fine-Tuned Large Language Model for Multi-label Social Media Text Classification in Disaster Informatics
In the field of crisis/disaster informatics, social media is increasingly being used for improving situational awareness to inform response and relief efforts. Efficient and accurate text classification tools have been a focal area of investigation in crisis informatics. However, current methods mostly rely on single-label text classification models, which fails to capture different insights embedded in dynamic and multifaceted disaster-related social media data. This study introduces a novel approach to disaster text classification by enhancing a pre-trained Large Language Model (LLM) through instruction fine-tuning targeted for multi-label classification of disaster-related tweets. Our methodology involves creating a comprehensive instruction dataset from disaster-related tweets, which is then used to fine-tune an open-source LLM, thereby embedding it with disaster-specific knowledge. This fine-tuned model can classify multiple aspects of disaster-related information simultaneously, such as the type of event, informativeness, and involvement of human aid, significantly improving the utility of social media data for situational awareness in disasters. The results demonstrate that this approach enhances the categorization of critical information from social media posts, thereby facilitating a more effective deployment for situational awareness during emergencies. This research paves the way for more advanced, adaptable, and robust disaster management tools, leveraging the capabilities of LLMs to improve real-time situational awareness and response strategies in disaster scenarios.
comment: Relevant source code and data is available: https://github.com/KaiYin97/CrsisLLM
♻ ☆ A General Framework for Inference-time Scaling and Steering of Diffusion Models
Diffusion models produce impressive results in modalities ranging from images and video to protein design and text. However, generating samples with user-specified properties remains a challenge. Recent research proposes fine-tuning models to maximize rewards that capture desired properties, but these methods require expensive training and are prone to mode collapse. In this work, we propose Feynman Kac (FK) steering, an inference-time framework for steering diffusion models with reward functions. FK steering works by sampling a system of multiple interacting diffusion processes, called particles, and resampling particles at intermediate steps based on scores computed using functions called potentials. Potentials are defined using rewards for intermediate states and are selected such that a high value indicates that the particle will yield a high-reward sample. We explore various choices of potentials, intermediate rewards, and samplers. We evaluate FK steering on text-to-image and text diffusion models. For steering text-to-image models with a human preference reward, we find that FK steering a 0.8B parameter model outperforms a 2.6B parameter fine-tuned model on prompt fidelity, with faster sampling and no training. For steering text diffusion models with rewards for text quality and specific text attributes, we find that FK steering generates lower perplexity, more linguistically acceptable outputs and enables gradient-free control of attributes like toxicity. Our results demonstrate that inference-time scaling and steering of diffusion models, even with off-the-shelf rewards, can provide significant sample quality gains and controllability benefits. Code is available at https://github.com/zacharyhorvitz/Fk-Diffusion-Steering .
♻ ☆ Surveying Attitudinal Alignment Between Large Language Models Vs. Humans Towards 17 Sustainable Development Goals
Large Language Models (LLMs) have emerged as potent tools for advancing the United Nations' Sustainable Development Goals (SDGs). However, the attitudinal disparities between LLMs and humans towards these goals can pose significant challenges. This study conducts a comprehensive review and analysis of the existing literature on the attitudes of LLMs towards the 17 SDGs, emphasizing the comparison between their attitudes and support for each goal and those of humans. We examine the potential disparities, primarily focusing on aspects such as understanding and emotions, cultural and regional differences, task objective variations, and factors considered in the decision-making process. These disparities arise from the underrepresentation and imbalance in LLM training data, historical biases, quality issues, lack of contextual understanding, and skewed ethical values reflected. The study also investigates the risks and harms that may arise from neglecting the attitudes of LLMs towards the SDGs, including the exacerbation of social inequalities, racial discrimination, environmental destruction, and resource wastage. To address these challenges, we propose strategies and recommendations to guide and regulate the application of LLMs, ensuring their alignment with the principles and goals of the SDGs, and therefore creating a more just, inclusive, and sustainable future.
♻ ☆ PeFoMed: Parameter Efficient Fine-tuning of Multimodal Large Language Models for Medical Imaging
Multimodal large language models (MLLMs) represent an evolutionary expansion in the capabilities of traditional large language models, enabling them to tackle challenges that surpass the scope of purely text-based applications. It leverages the knowledge previously encoded within these language models, thereby enhancing their applicability and functionality in the reign of multimodal contexts. Recent works investigate the adaptation of MLLMs as a universal solution to address medical multi-modal problems as a generative task. In this paper, we propose a parameter efficient framework for fine-tuning MLLMs, specifically validated on medical visual question answering (Med-VQA) and medical report generation (MRG) tasks, using public benchmark datasets. We also introduce an evaluation metric using the 5-point Likert scale and its weighted average value to measure the quality of the generated reports for MRG tasks, where the scale ratings are labelled by both humans manually and the GPT-4 model. We further assess the consistency of performance metrics across traditional measures, GPT-4, and human ratings for both VQA and MRG tasks. The results indicate that semantic similarity assessments using GPT-4 align closely with human annotators and provide greater stability, yet they reveal a discrepancy when compared to conventional lexical similarity measurements. This questions the reliability of lexical similarity metrics for evaluating the performance of generative models in Med-VQA and report generation tasks. Besides, our fine-tuned model significantly outperforms GPT-4v. This indicates that without additional fine-tuning, multi-modal models like GPT-4v do not perform effectively on medical imaging tasks. The code will be available here: https://github.com/jinlHe/PeFoMed.
comment: 12 pages, 8 figures, 12 tables
♻ ☆ BLEnD: A Benchmark for LLMs on Everyday Knowledge in Diverse Cultures and Languages NeurIPS 2024
Large language models (LLMs) often lack culture-specific knowledge of daily life, especially across diverse regions and non-English languages. Existing benchmarks for evaluating LLMs' cultural sensitivities are limited to a single language or collected from online sources such as Wikipedia, which do not reflect the mundane everyday lifestyles of diverse regions. That is, information about the food people eat for their birthday celebrations, spices they typically use, musical instruments youngsters play, or the sports they practice in school is common cultural knowledge but uncommon in easily collected online sources, especially for underrepresented cultures. To address this issue, we introduce BLEnD, a hand-crafted benchmark designed to evaluate LLMs' everyday knowledge across diverse cultures and languages. BLEnD comprises 52.6k question-answer pairs from 16 countries/regions, in 13 different languages, including low-resource ones such as Amharic, Assamese, Azerbaijani, Hausa, and Sundanese. We construct the benchmark to include two formats of questions: short-answer and multiple-choice. We show that LLMs perform better for cultures that are highly represented online, with a maximum 57.34% difference in GPT-4, the best-performing model, in the short-answer format. For cultures represented by mid-to-high-resource languages, LLMs perform better in their local languages, but for cultures represented by low-resource languages, LLMs perform better in English than the local languages. We make our dataset publicly available at: https://github.com/nlee0212/BLEnD.
comment: Accepted to NeurIPS 2024 Datasets & Benchmark Track
♻ ☆ Entailed Between the Lines: Incorporating Implication into NLI
Much of human communication depends on implication, conveying meaning beyond literal words to express a wider range of thoughts, intentions, and feelings. For models to better understand and facilitate human communication, they must be responsive to the text's implicit meaning. We focus on Natural Language Inference (NLI), a core tool for many language tasks, and find that state-of-the-art NLI models and datasets struggle to recognize a range of cases where entailment is implied, rather than explicit from the text. We formalize implied entailment as an extension of the NLI task and introduce the Implied NLI dataset (INLI) to help today's LLMs both recognize a broader variety of implied entailments and to distinguish between implicit and explicit entailment. We show how LLMs fine-tuned on INLI understand implied entailment and can generalize this understanding across datasets and domains.
Machine Learning 155
☆ SRE-Conv: Symmetric Rotation Equivariant Convolution for Biomedical Image Classification
Convolutional neural networks (CNNs) are essential tools for computer vision tasks, but they lack traditionally desired properties of extracted features that could further improve model performance, e.g., rotational equivariance. Such properties are ubiquitous in biomedical images, which often lack explicit orientation. While current work largely relies on data augmentation or explicit modules to capture orientation information, this comes at the expense of increased training costs or ineffective approximations of the desired equivariance. To overcome these challenges, we propose a novel and efficient implementation of the Symmetric Rotation-Equivariant (SRE) Convolution (SRE-Conv) kernel, designed to learn rotation-invariant features while simultaneously compressing the model size. The SRE-Conv kernel can easily be incorporated into any CNN backbone. We validate the ability of a deep SRE-CNN to capture equivariance to rotation using the public MedMNISTv2 dataset (16 total tasks). SRE-Conv-CNN demonstrated improved rotated image classification performance accuracy on all 16 test datasets in both 2D and 3D images, all while increasing efficiency with fewer parameters and reduced memory footprint. The code is available at https://github.com/XYPB/SRE-Conv.
comment: Accepted by IEEE ISBI 2025 4-page paper
☆ FAST: Efficient Action Tokenization for Vision-Language-Action Models
Autoregressive sequence models, such as Transformer-based vision-language action (VLA) policies, can be tremendously effective for capturing complex and generalizable robotic behaviors. However, such models require us to choose a tokenization of our continuous action signals, which determines how the discrete symbols predicted by the model map to continuous robot actions. We find that current approaches for robot action tokenization, based on simple per-dimension, per-timestep binning schemes, typically perform poorly when learning dexterous skills from high-frequency robot data. To address this challenge, we propose a new compression-based tokenization scheme for robot actions, based on the discrete cosine transform. Our tokenization approach, Frequency-space Action Sequence Tokenization (FAST), enables us to train autoregressive VLAs for highly dexterous and high-frequency tasks where standard discretization methods fail completely. Based on FAST, we release FAST+, a universal robot action tokenizer, trained on 1M real robot action trajectories. It can be used as a black-box tokenizer for a wide range of robot action sequences, with diverse action spaces and control frequencies. Finally, we show that, when combined with the pi0 VLA, our method can scale to training on 10k hours of robot data and match the performance of diffusion VLAs, while reducing training time by up to 5x.
comment: Website: https://www.pi.website/research/fast
☆ Suggesting Code Edits in Interactive Machine Learning Notebooks Using Large Language Models
Machine learning developers frequently use interactive computational notebooks, such as Jupyter notebooks, to host code for data processing and model training. Jupyter notebooks provide a convenient tool for writing machine learning pipelines and interactively observing outputs, however, maintaining Jupyter notebooks, e.g., to add new features or fix bugs, can be challenging due to the length and complexity of the notebooks. Moreover, there is no existing benchmark related to developer edits on Jupyter notebooks. To address this, we present the first dataset of 48,398 Jupyter notebook edits derived from 20,095 revisions of 792 machine learning repositories on GitHub, and perform the first study of the using LLMs to predict code edits in Jupyter notebooks. Our dataset captures granular details of cell-level and line-level modifications, offering a foundation for understanding real-world maintenance patterns in machine learning workflows. We observed that the edits on Jupyter notebooks are highly localized, with changes averaging only 166 lines of code in repositories. While larger models outperform smaller counterparts in code editing, all models have low accuracy on our dataset even after finetuning, demonstrating the complexity of real-world machine learning maintenance tasks. Our findings emphasize the critical role of contextual information in improving model performance and point toward promising avenues for advancing large language models' capabilities in engineering machine learning code.
☆ Random Subspace Cubic-Regularization Methods, with Applications to Low-Rank Functions
We propose and analyze random subspace variants of the second-order Adaptive Regularization using Cubics (ARC) algorithm. These methods iteratively restrict the search space to some random subspace of the parameters, constructing and minimizing a local model only within this subspace. Thus, our variants only require access to (small-dimensional) projections of first- and second-order problem derivatives and calculate a reduced step inexpensively. Under suitable assumptions, the ensuing methods maintain the optimal first-order, and second-order, global rates of convergence of (full-dimensional) cubic regularization, while showing improved scalability both theoretically and numerically, particularly when applied to low-rank functions. When applied to the latter, our adaptive variant naturally adapts the subspace size to the true rank of the function, without knowing it a priori.
☆ Predictions as Surrogates: Revisiting Surrogate Outcomes in the Age of AI
We establish a formal connection between the decades-old surrogate outcome model in biostatistics and economics and the emerging field of prediction-powered inference (PPI). The connection treats predictions from pre-trained models, prevalent in the age of AI, as cost-effective surrogates for expensive outcomes. Building on the surrogate outcomes literature, we develop recalibrated prediction-powered inference, a more efficient approach to statistical inference than existing PPI proposals. Our method departs from the existing proposals by using flexible machine learning techniques to learn the optimal ``imputed loss'' through a step we call recalibration. Importantly, the method always improves upon the estimator that relies solely on the data with available true outcomes, even when the optimal imputed loss is estimated imperfectly, and it achieves the smallest asymptotic variance among PPI estimators if the estimate is consistent. Computationally, our optimization objective is convex whenever the loss function that defines the target parameter is convex. We further analyze the benefits of recalibration, both theoretically and numerically, in several common scenarios where machine learning predictions systematically deviate from the outcome of interest. We demonstrate significant gains in effective sample size over existing PPI proposals via three applications leveraging state-of-the-art machine learning/AI models.
☆ Generating particle physics Lagrangians with transformers
In physics, Lagrangians provide a systematic way to describe laws governing physical systems. In the context of particle physics, they encode the interactions and behavior of the fundamental building blocks of our universe. By treating Lagrangians as complex, rule-based constructs similar to linguistic expressions, we trained a transformer model -- proven to be effective in natural language tasks -- to predict the Lagrangian corresponding to a given list of particles. We report on the transformer's performance in constructing Lagrangians respecting the Standard Model $\mathrm{SU}(3)\times \mathrm{SU}(2)\times \mathrm{U}(1)$ gauge symmetries. The resulting model is shown to achieve high accuracies (over 90\%) with Lagrangians up to six matter fields, with the capacity to generalize beyond the training distribution, albeit within architectural constraints. We show through an analysis of input embeddings that the model has internalized concepts such as group representations and conjugation operations as it learned to generate Lagrangians. We make the model and training datasets available to the community. An interactive demonstration can be found at: \url{https://huggingface.co/spaces/JoseEliel/generate-lagrangians}.
comment: 32 pages, 11 figues, 18 tables
☆ Attention based Bidirectional GRU hybrid model for inappropriate content detection in Urdu language
With the increased use of the internet and social networks for online discussions, the spread of toxic and inappropriate content on social networking sites has also increased. Several studies have been conducted in different languages. However, there is less work done for South Asian languages for inappropriate content identification using deep learning techniques. In Urdu language, the spellings are not unique, and people write different common spellings for the same word, while mixing it other languages, like English in the text makes it more challenging, and limited research work is available to process such language with the finest algorithms. The use of attention layer with a deep learning model can help handling the long-term dependencies and increase its efficiency . To explore the effects of the attention layer, this study proposes attention-based Bidirectional GRU hybrid model for identifying inappropriate content in Urdu Unicode text language. Four different baseline deep learning models; LSTM, Bi-LSTM, GRU, and TCN, are used to compare the performance of the proposed model. The results of these models were compared based on evaluation metrics, dataset size, and impact of the word embedding layer. The pre-trained Urdu word2Vec embeddings were utilized for our case. Our proposed model BiGRU-A outperformed all other baseline models by yielding 84\% accuracy without using pre-trained word2Vec layer. From our experiments, we have established that the attention layer improves the model's efficiency, and pre-trained word2Vec embedding does not work well with an inappropriate content dataset.
☆ Practical Continual Forgetting for Pre-trained Vision Models
For privacy and security concerns, the need to erase unwanted information from pre-trained vision models is becoming evident nowadays. In real-world scenarios, erasure requests originate at any time from both users and model owners, and these requests usually form a sequence. Therefore, under such a setting, selective information is expected to be continuously removed from a pre-trained model while maintaining the rest. We define this problem as continual forgetting and identify three key challenges. (i) For unwanted knowledge, efficient and effective deleting is crucial. (ii) For remaining knowledge, the impact brought by the forgetting procedure should be minimal. (iii) In real-world scenarios, the training samples may be scarce or partially missing during the process of forgetting. To address them, we first propose Group Sparse LoRA (GS-LoRA). Specifically, towards (i), we introduce LoRA modules to fine-tune the FFN layers in Transformer blocks for each forgetting task independently, and towards (ii), a simple group sparse regularization is adopted, enabling automatic selection of specific LoRA groups and zeroing out the others. To further extend GS-LoRA to more practical scenarios, we incorporate prototype information as additional supervision and introduce a more practical approach, GS-LoRA++. For each forgotten class, we move the logits away from its original prototype. For the remaining classes, we pull the logits closer to their respective prototypes. We conduct extensive experiments on face recognition, object detection and image classification and demonstrate that our method manages to forget specific classes with minimal impact on other classes. Codes have been released on https://github.com/bjzhb666/GS-LoRA.
☆ Cueless EEG imagined speech for subject identification: dataset and benchmarks
Electroencephalogram (EEG) signals have emerged as a promising modality for biometric identification. While previous studies have explored the use of imagined speech with semantically meaningful words for subject identification, most have relied on additional visual or auditory cues. In this study, we introduce a cueless EEG-based imagined speech paradigm, where subjects imagine the pronunciation of semantically meaningful words without any external cues. This innovative approach addresses the limitations of prior methods by requiring subjects to select and imagine words from a predefined list naturally. The dataset comprises over 4,350 trials from 11 subjects across five sessions. We assess a variety of classification methods, including traditional machine learning techniques such as Support Vector Machines (SVM) and XGBoost, as well as time-series foundation models and deep learning architectures specifically designed for EEG classification, such as EEG Conformer and Shallow ConvNet. A session-based hold-out validation strategy was employed to ensure reliable evaluation and prevent data leakage. Our results demonstrate outstanding classification accuracy, reaching 97.93%. These findings highlight the potential of cueless EEG paradigms for secure and reliable subject identification in real-world applications, such as brain-computer interfaces (BCIs).
☆ A Near-optimal Algorithm for Learning Margin Halfspaces with Massart Noise
We study the problem of PAC learning $\gamma$-margin halfspaces in the presence of Massart noise. Without computational considerations, the sample complexity of this learning problem is known to be $\widetilde{\Theta}(1/(\gamma^2 \epsilon))$. Prior computationally efficient algorithms for the problem incur sample complexity $\tilde{O}(1/(\gamma^4 \epsilon^3))$ and achieve 0-1 error of $\eta+\epsilon$, where $\eta<1/2$ is the upper bound on the noise rate. Recent work gave evidence of an information-computation tradeoff, suggesting that a quadratic dependence on $1/\epsilon$ is required for computationally efficient algorithms. Our main result is a computationally efficient learner with sample complexity $\widetilde{\Theta}(1/(\gamma^2 \epsilon^2))$, nearly matching this lower bound. In addition, our algorithm is simple and practical, relying on online SGD on a carefully selected sequence of convex losses.
☆ U-Fair: Uncertainty-based Multimodal Multitask Learning for Fairer Depression Detection ML4H
Machine learning bias in mental health is becoming an increasingly pertinent challenge. Despite promising efforts indicating that multitask approaches often work better than unitask approaches, there is minimal work investigating the impact of multitask learning on performance and fairness in depression detection nor leveraged it to achieve fairer prediction outcomes. In this work, we undertake a systematic investigation of using a multitask approach to improve performance and fairness for depression detection. We propose a novel gender-based task-reweighting method using uncertainty grounded in how the PHQ-8 questionnaire is structured. Our results indicate that, although a multitask approach improves performance and fairness compared to a unitask approach, the results are not always consistent and we see evidence of negative transfer and a reduction in the Pareto frontier, which is concerning given the high-stake healthcare setting. Our proposed approach of gender-based reweighting with uncertainty improves performance and fairness and alleviates both challenges to a certain extent. Our findings on each PHQ-8 subitem task difficulty are also in agreement with the largest study conducted on the PHQ-8 subitem discrimination capacity, thus providing the very first tangible evidence linking ML findings with large-scale empirical population studies conducted on the PHQ-8.
comment: To appear at the Proceedings of Machine Learning Research 259, 1-14, 2024 as part of the Machine Learning for Health (ML4H) Symposium 2024
☆ Reward-Guided Controlled Generation for Inference-Time Alignment in Diffusion Models: Tutorial and Review
This tutorial provides an in-depth guide on inference-time guidance and alignment methods for optimizing downstream reward functions in diffusion models. While diffusion models are renowned for their generative modeling capabilities, practical applications in fields such as biology often require sample generation that maximizes specific metrics (e.g., stability, affinity in proteins, closeness to target structures). In these scenarios, diffusion models can be adapted not only to generate realistic samples but also to explicitly maximize desired measures at inference time without fine-tuning. This tutorial explores the foundational aspects of such inference-time algorithms. We review these methods from a unified perspective, demonstrating that current techniques -- such as Sequential Monte Carlo (SMC)-based guidance, value-based sampling, and classifier guidance -- aim to approximate soft optimal denoising processes (a.k.a. policies in RL) that combine pre-trained denoising processes with value functions serving as look-ahead functions that predict from intermediate states to terminal rewards. Within this framework, we present several novel algorithms not yet covered in the literature. Furthermore, we discuss (1) fine-tuning methods combined with inference-time techniques, (2) inference-time algorithms based on search algorithms such as Monte Carlo tree search, which have received limited attention in current research, and (3) connections between inference-time algorithms in language models and diffusion models. The code of this tutorial on protein design is available at https://github.com/masa-ue/AlignInversePro
comment: We plan to add more content/codes. Please let us know if there are any comments
☆ Rough kernel hedging
Building on the functional-analytic framework of operator-valued kernels and un-truncated signature kernels, we propose a scalable, provably convergent signature-based algorithm for a broad class of high-dimensional, path-dependent hedging problems. We make minimal assumptions about market dynamics by modelling them as general geometric rough paths, yielding a fully model-free approach. Furthermore, through a representer theorem, we provide theoretical guarantees on the existence and uniqueness of a global minimum for the resulting optimization problem and derive an analytic solution under highly general loss functions. Similar to the popular deep hedging approach, but in a more rigorous fashion, our method can also incorporate additional features via the underlying operator-valued kernel, such as trading signals, news analytics, and past hedging decisions, closely aligning with true machine-learning practice.
☆ Fokker-Planck to Callan-Symanzik: evolution of weight matrices under training
The dynamical evolution of a neural network during training has been an incredibly fascinating subject of study. First principal derivation of generic evolution of variables in statistical physics systems has proved useful when used to describe training dynamics conceptually, which in practice means numerically solving equations such as Fokker-Planck equation. Simulating entire networks inevitably runs into the curse of dimensionality. In this paper, we utilize Fokker-Planck to simulate the probability density evolution of individual weight matrices in the bottleneck layers of a simple 2-bottleneck-layered auto-encoder and compare the theoretical evolutions against the empirical ones by examining the output data distributions. We also derive physically relevant partial differential equations such as Callan-Symanzik and Kardar-Parisi-Zhang equations from the dynamical equation we have.
comment: 8 pages, 9 figures
☆ A Survey of Research in Large Language Models for Electronic Design Automation
Within the rapidly evolving domain of Electronic Design Automation (EDA), Large Language Models (LLMs) have emerged as transformative technologies, offering unprecedented capabilities for optimizing and automating various aspects of electronic design. This survey provides a comprehensive exploration of LLM applications in EDA, focusing on advancements in model architectures, the implications of varying model sizes, and innovative customization techniques that enable tailored analytical insights. By examining the intersection of LLM capabilities and EDA requirements, the paper highlights the significant impact these models have on extracting nuanced understandings from complex datasets. Furthermore, it addresses the challenges and opportunities in integrating LLMs into EDA workflows, paving the way for future research and application in this dynamic field. Through this detailed analysis, the survey aims to offer valuable insights to professionals in the EDA industry, AI researchers, and anyone interested in the convergence of advanced AI technologies and electronic design.
comment: 21 pages, 2 figures, 3 tables, accepted by TODAES
☆ LLM-Based Routing in Mixture of Experts: A Novel Framework for Trading AAAI 2025
Recent advances in deep learning and large language models (LLMs) have facilitated the deployment of the mixture-of-experts (MoE) mechanism in the stock investment domain. While these models have demonstrated promising trading performance, they are often unimodal, neglecting the wealth of information available in other modalities, such as textual data. Moreover, the traditional neural network-based router selection mechanism fails to consider contextual and real-world nuances, resulting in suboptimal expert selection. To address these limitations, we propose LLMoE, a novel framework that employs LLMs as the router within the MoE architecture. Specifically, we replace the conventional neural network-based router with LLMs, leveraging their extensive world knowledge and reasoning capabilities to select experts based on historical price data and stock news. This approach provides a more effective and interpretable selection mechanism. Our experiments on multimodal real-world stock datasets demonstrate that LLMoE outperforms state-of-the-art MoE models and other deep neural network approaches. Additionally, the flexible architecture of LLMoE allows for easy adaptation to various downstream tasks.
comment: Accepted by AAAI 2025 Workshop on AI for Social Impact - Bridging Innovations in Finance, Social Media, and Crime Prevention
☆ Empowering Large Language Models in Wireless Communication: A Novel Dataset and Fine-Tuning Framework
In this work, we develop a specialized dataset aimed at enhancing the evaluation and fine-tuning of large language models (LLMs) specifically for wireless communication applications. The dataset includes a diverse set of multi-hop questions, including true/false and multiple-choice types, spanning varying difficulty levels from easy to hard. By utilizing advanced language models for entity extraction and question generation, rigorous data curation processes are employed to maintain high quality and relevance. Additionally, we introduce a Pointwise V-Information (PVI) based fine-tuning method, providing a detailed theoretical analysis and justification for its use in quantifying the information content of training data with 2.24\% and 1.31\% performance boost for different models compared to baselines, respectively. To demonstrate the effectiveness of the fine-tuned models with the proposed methodologies on practical tasks, we also consider different tasks, including summarizing optimization problems from technical papers and solving the mathematical problems related to non-orthogonal multiple access (NOMA), which are generated by using the proposed multi-agent framework. Simulation results show significant performance gain in summarization tasks with 20.9\% in the ROUGE-L metrics. We also study the scaling laws of fine-tuning LLMs and the challenges LLMs face in the field of wireless communications, offering insights into their adaptation to wireless communication tasks. This dataset and fine-tuning methodology aim to enhance the training and evaluation of LLMs, contributing to advancements in LLMs for wireless communication research and applications.
comment: 13 pages, 13 figure, journal
☆ Weight for Robustness: A Comprehensive Approach towards Optimal Fault-Tolerant Asynchronous ML
We address the challenges of Byzantine-robust training in asynchronous distributed machine learning systems, aiming to enhance efficiency amid massive parallelization and heterogeneous computing resources. Asynchronous systems, marked by independently operating workers and intermittent updates, uniquely struggle with maintaining integrity against Byzantine failures, which encompass malicious or erroneous actions that disrupt learning. The inherent delays in such settings not only introduce additional bias to the system but also obscure the disruptions caused by Byzantine faults. To tackle these issues, we adapt the Byzantine framework to asynchronous dynamics by introducing a novel weighted robust aggregation framework. This allows for the extension of robust aggregators and a recent meta-aggregator to their weighted versions, mitigating the effects of delayed updates. By further incorporating a recent variance-reduction technique, we achieve an optimal convergence rate for the first time in an asynchronous Byzantine environment. Our methodology is rigorously validated through empirical and theoretical analysis, demonstrating its effectiveness in enhancing fault tolerance and optimizing performance in asynchronous ML systems.
☆ Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment
Recent advances in large language models (LLMs) have demonstrated significant progress in performing complex tasks. While Reinforcement Learning from Human Feedback (RLHF) has been effective in aligning LLMs with human preferences, it is susceptible to spurious correlations in reward modeling. Consequently, it often introduces biases-such as length bias, sycophancy, conceptual bias, and discrimination that hinder the model's ability to capture true causal relationships. To address this, we propose a novel causal reward modeling approach that integrates causal inference to mitigate these spurious correlations. Our method enforces counterfactual invariance, ensuring reward predictions remain consistent when irrelevant variables are altered. Through experiments on both synthetic and real-world datasets, we show that our approach mitigates various types of spurious correlations effectively, resulting in more reliable and fair alignment of LLMs with human preferences. As a drop-in enhancement to the existing RLHF workflow, our causal reward modeling provides a practical way to improve the trustworthiness and fairness of LLM finetuning.
☆ ARMAX identification of low rank graphical models
In large-scale systems, complex internal relationships are often present. Such interconnected systems can be effectively described by low rank stochastic processes. When identifying a predictive model of low rank processes from sampling data, the rank-deficient property of spectral densities is often obscured by the inevitable measurement noise in practice. However, existing low rank identification approaches often did not take noise into explicit consideration, leading to non-negligible inaccuracies even under weak noise. In this paper, we address the identification issue of low rank processes under measurement noise. We find that the noisy measurement model admits a sparse plus low rank structure in latent-variable graphical models. Specifically, we first decompose the problem into a maximum entropy covariance extension problem, and a low rank graphical estimation problem based on an autoregressive moving-average with exogenous input (ARMAX) model. To identify the ARMAX low rank graphical models, we propose an estimation approach based on maximum likelihood. The identifiability and consistency of this approach are proven under certain conditions. Simulation results confirm the reliable performance of the entire algorithm in both the parameter estimation and noisy data filtering.
☆ EVaDE : Event-Based Variational Thompson Sampling for Model-Based Reinforcement Learning
Posterior Sampling for Reinforcement Learning (PSRL) is a well-known algorithm that augments model-based reinforcement learning (MBRL) algorithms with Thompson sampling. PSRL maintains posterior distributions of the environment transition dynamics and the reward function, which are intractable for tasks with high-dimensional state and action spaces. Recent works show that dropout, used in conjunction with neural networks, induces variational distributions that can approximate these posteriors. In this paper, we propose Event-based Variational Distributions for Exploration (EVaDE), which are variational distributions that are useful for MBRL, especially when the underlying domain is object-based. We leverage the general domain knowledge of object-based domains to design three types of event-based convolutional layers to direct exploration. These layers rely on Gaussian dropouts and are inserted between the layers of the deep neural network model to help facilitate variational Thompson sampling. We empirically show the effectiveness of EVaDE-equipped Simulated Policy Learning (EVaDE-SimPLe) on the 100K Atari game suite.
☆ Adversarial-Ensemble Kolmogorov Arnold Networks for Enhancing Indoor Wi-Fi Positioning: A Defensive Approach Against Spoofing and Signal Manipulation Attacks
The research presents a study on enhancing the robustness of Wi-Fi-based indoor positioning systems against adversarial attacks. The goal is to improve the positioning accuracy and resilience of these systems under two attack scenarios: Wi-Fi Spoofing and Signal Strength Manipulation. Three models are developed and evaluated: a baseline model (M_Base), an adversarially trained robust model (M_Rob), and an ensemble model (M_Ens). All models utilize a Kolmogorov-Arnold Network (KAN) architecture. The robust model is trained with adversarially perturbed data, while the ensemble model combines predictions from both the base and robust models. Experimental results show that the robust model reduces positioning error by approximately 10% compared to the baseline, achieving 2.03 meters error under Wi-Fi spoofing and 2.00 meters under signal strength manipulation. The ensemble model further outperforms with errors of 2.01 meters and 1.975 meters for the respective attack types. This analysis highlights the effectiveness of adversarial training techniques in mitigating attack impacts. The findings underscore the importance of considering adversarial scenarios in developing indoor positioning systems, as improved resilience can significantly enhance the accuracy and reliability of such systems in mission-critical environments.
☆ Reducing the Sensitivity of Neural Physics Simulators to Mesh Topology via Pretraining
Meshes are used to represent complex objects in high fidelity physics simulators across a variety of domains, such as radar sensing and aerodynamics. There is growing interest in using neural networks to accelerate physics simulations, and also a growing body of work on applying neural networks directly to irregular mesh data. Since multiple mesh topologies can represent the same object, mesh augmentation is typically required to handle topological variation when training neural networks. Due to the sensitivity of physics simulators to small changes in mesh shape, it is challenging to use these augmentations when training neural network-based physics simulators. In this work, we show that variations in mesh topology can significantly reduce the performance of neural network simulators. We evaluate whether pretraining can be used to address this issue, and find that employing an established autoencoder pretraining technique with graph embedding models reduces the sensitivity of neural network simulators to variations in mesh topology. Finally, we highlight future research directions that may further reduce neural simulator sensitivity to mesh topology.
comment: 5 pages, 3 figures
☆ IFRA: a machine learning-based Instrumented Fall Risk Assessment Scale derived from Instrumented Timed Up and Go test in stroke patients
Effective fall risk assessment is critical for post-stroke patients. The present study proposes a novel, data-informed fall risk assessment method based on the instrumented Timed Up and Go (ITUG) test data, bringing in many mobility measures that traditional clinical scales fail to capture. IFRA, which stands for Instrumented Fall Risk Assessment, has been developed using a two-step process: first, features with the highest predictive power among those collected in a ITUG test have been identified using machine learning techniques; then, a strategy is proposed to stratify patients into low, medium, or high-risk strata. The dataset used in our analysis consists of 142 participants, out of which 93 were used for training (15 synthetically generated), 17 for validation and 32 to test the resulting IFRA scale (22 non-fallers and 10 fallers). Features considered in the IFRA scale include gait speed, vertical acceleration during sit-to-walk transition, and turning angular velocity, which align well with established literature on the risk of fall in neurological patients. In a comparison with traditional clinical scales such as the traditional Timed Up & Go and the Mini-BESTest, IFRA demonstrates competitive performance, being the only scale to correctly assign more than half of the fallers to the high-risk stratum (Fischer's Exact test p = 0.004). Despite the dataset's limited size, this is the first proof-of-concept study to pave the way for future evidence regarding the use of IFRA tool for continuous patient monitoring and fall prevention both in clinical stroke rehabilitation and at home post-discharge.
comment: 26 pages, 2 figures, submitted for review dec 2024
☆ Metrics for Inter-Dataset Similarity with Example Applications in Synthetic Data and Feature Selection Evaluation -- Extended Version SDM
Measuring inter-dataset similarity is an important task in machine learning and data mining with various use cases and applications. Existing methods for measuring inter-dataset similarity are computationally expensive, limited, or sensitive to different entities and non-trivial choices for parameters. They also lack a holistic perspective on the entire dataset. In this paper, we propose two novel metrics for measuring inter-dataset similarity. We discuss the mathematical foundation and the theoretical basis of our proposed metrics. We demonstrate the effectiveness of the proposed metrics by investigating two applications in the evaluation of synthetic data and in the evaluation of feature selection methods. The theoretical and empirical studies conducted in this paper illustrate the effectiveness of the proposed metrics.
comment: This is the extended version of a paper accepted at 2025 SIAM International Conference on Data Mining (SDM)
☆ Atleus: Accelerating Transformers on the Edge Enabled by 3D Heterogeneous Manycore Architectures
Transformer architectures have become the standard neural network model for various machine learning applications including natural language processing and computer vision. However, the compute and memory requirements introduced by transformer models make them challenging to adopt for edge applications. Furthermore, fine-tuning pre-trained transformers (e.g., foundation models) is a common task to enhance the model's predictive performance on specific tasks/applications. Existing transformer accelerators are oblivious to complexities introduced by fine-tuning. In this paper, we propose the design of a three-dimensional (3D) heterogeneous architecture referred to as Atleus that incorporates heterogeneous computing resources specifically optimized to accelerate transformer models for the dual purposes of fine-tuning and inference. Specifically, Atleus utilizes non-volatile memory and systolic array for accelerating transformer computational kernels using an integrated 3D platform. Moreover, we design a suitable NoC to achieve high performance and energy efficiency. Finally, Atleus adopts an effective quantization scheme to support model compression. Experimental results demonstrate that Atleus outperforms existing state-of-the-art by up to 56x and 64.5x in terms of performance and energy efficiency respectively
comment: Accepted for Publication in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD)
☆ Sequential PatchCore: Anomaly Detection for Surface Inspection using Synthetic Impurities
The appearance of surface impurities (e.g., water stains, fingerprints, stickers) is an often-mentioned issue that causes degradation of automated visual inspection systems. At the same time, synthetic data generation techniques for visual surface inspection have focused primarily on generating perfect examples and defects, disregarding impurities. This study highlights the importance of considering impurities when generating synthetic data. We introduce a procedural method to include photorealistic water stains in synthetic data. The synthetic datasets are generated to correspond to real datasets and are further used to train an anomaly detection model and investigate the influence of water stains. The high-resolution images used for surface inspection lead to memory bottlenecks during anomaly detection training. To address this, we introduce Sequential PatchCore - a method to build coresets sequentially and make training on large images using consumer-grade hardware tractable. This allows us to perform transfer learning using coresets pre-trained on different dataset versions. Our results show the benefits of using synthetic data for pre-training an explicit coreset anomaly model and the extended performance benefits of finetuning the coreset using real data. We observed how the impurities and labelling ambiguity lower the model performance and have additionally reported the defect-wise recall to provide an industrially relevant perspective on model performance.
☆ Towards Spectral Convergence of Locally Linear Embedding on Manifolds with Boundary
We study the eigenvalues and eigenfunctions of a differential operator that governs the asymptotic behavior of the unsupervised learning algorithm known as Locally Linear Embedding when a large data set is sampled from an interval or disc. In particular, the differential operator is of second order, mixed-type, and degenerates near the boundary. We show that a natural regularity condition on the eigenfunctions imposes a consistent boundary condition and use the Frobenius method to estimate pointwise behavior. We then determine the limiting sequence of eigenvalues analytically and compare them to numerical predictions. Finally, we propose a variational framework for determining eigenvalues on other compact manifolds.
comment: 26 pages, 7 figures; the author welcomes all comments
☆ MatrixNet: Learning over symmetry groups using learned group representations NeurIPS 2024
Group theory has been used in machine learning to provide a theoretically grounded approach for incorporating known symmetry transformations in tasks from robotics to protein modeling. In these applications, equivariant neural networks use known symmetry groups with predefined representations to learn over geometric input data. We propose MatrixNet, a neural network architecture that learns matrix representations of group element inputs instead of using predefined representations. MatrixNet achieves higher sample efficiency and generalization over several standard baselines in prediction tasks over the several finite groups and the Artin braid group. We also show that MatrixNet respects group relations allowing generalization to group elements of greater word length than in the training set.
comment: NeurIPS 2024
☆ Overshoot: Taking advantage of future gradients in momentum-based stochastic optimization
Overshoot is a novel, momentum-based stochastic gradient descent optimization method designed to enhance performance beyond standard and Nesterov's momentum. In conventional momentum methods, gradients from previous steps are aggregated with the gradient at current model weights before taking a step and updating the model. Rather than calculating gradient at the current model weights, Overshoot calculates the gradient at model weights shifted in the direction of the current momentum. This sacrifices the immediate benefit of using the gradient w.r.t. the exact model weights now, in favor of evaluating at a point, which will likely be more relevant for future updates. We show that incorporating this principle into momentum-based optimizers (SGD with momentum and Adam) results in faster convergence (saving on average at least 15% of steps). Overshoot consistently outperforms both standard and Nesterov's momentum across a wide range of tasks and integrates into popular momentum-based optimizers with zero memory and small computational overhead.
☆ Intra-day Solar and Power Forecast for Optimization of Intraday Market Participation
The prediction of solar irradiance enhances reliability in photovoltaic (PV) solar plant generation and grid integration. In Colombia, PV plants face penalties if energy production deviates beyond governmental thresholds from intraday market offers. This research employs Long Short-Term Memory (LSTM) and Bidirectional-LSTM (Bi-LSTM) models, utilizing meteorological data from a PV plant in El Paso, Cesar, Colombia, to predict solar irradiance with a 6-hour horizon and 10-minute resolution. While Bi-LSTM showed superior performance, the LSTM model achieved comparable results with significantly reduced training time (6 hours versus 18 hours), making it computationally advantageous. The LSTM predictions were averaged to create an hourly resolution model, evaluated using Mean Absolute Error, Root-Mean-Square Error, Normalized Root-Mean-Square Error, and Mean Absolute Percentage Error metrics. Comparison with the Global Forecast System (GFS) revealed similar performance, with both models effectively capturing daily solar irradiance patterns. The forecast model integrates with an Object-Oriented power production model, enabling accurate energy offers in the intraday market while minimizing penalty costs.
comment: 20 pages, 37 figures, 9 tables
☆ MOGNET: A Mux-residual quantized Network leveraging Online-Generated weights
This paper presents a compact model architecture called MOGNET, compatible with a resource-limited hardware. MOGNET uses a streamlined Convolutional factorization block based on a combination of 2 point-wise (1x1) convolutions with a group-wise convolution in-between. To further limit the overall model size and reduce the on-chip required memory, the second point-wise convolution's parameters are on-line generated by a Cellular Automaton structure. In addition, MOGNET enables the use of low-precision weights and activations, by taking advantage of a Multiplexer mechanism with a proper Bitshift rescaling for integrating residual paths without increasing the hardware-related complexity. To efficiently train this model we also introduce a novel weight ternarization method favoring the balance between quantized levels. Experimental results show that given tiny memory budget (sub-2Mb), MOGNET can achieve higher accuracy with a clear gap up to 1% at a similar or even lower model size compared to recent state-of-the-art methods.
comment: Published at IEEE AICAS 2022
☆ Confidence Estimation for Error Detection in Text-to-SQL Systems AAAI 2025
Text-to-SQL enables users to interact with databases through natural language, simplifying the retrieval and synthesis of information. Despite the success of large language models (LLMs) in converting natural language questions into SQL queries, their broader adoption is limited by two main challenges: achieving robust generalization across diverse queries and ensuring interpretative confidence in their predictions. To tackle these issues, our research investigates the integration of selective classifiers into Text-to-SQL systems. We analyse the trade-off between coverage and risk using entropy based confidence estimation with selective classifiers and assess its impact on the overall performance of Text-to-SQL models. Additionally, we explore the models' initial calibration and improve it with calibration techniques for better model alignment between confidence and accuracy. Our experimental results show that encoder-decoder T5 is better calibrated than in-context-learning GPT 4 and decoder-only Llama 3, thus the designated external entropy-based selective classifier has better performance. The study also reveal that, in terms of error detection, selective classifier with a higher probability detects errors associated with irrelevant questions rather than incorrect query generations.
comment: 15 pages, 11 figures, to be published in AAAI 2025 Proceedings
☆ Class Incremental Fault Diagnosis under Limited Fault Data via Supervised Contrastive Knowledge Distillation
Class-incremental fault diagnosis requires a model to adapt to new fault classes while retaining previous knowledge. However, limited research exists for imbalanced and long-tailed data. Extracting discriminative features from few-shot fault data is challenging, and adding new fault classes often demands costly model retraining. Moreover, incremental training of existing methods risks catastrophic forgetting, and severe class imbalance can bias the model's decisions toward normal classes. To tackle these issues, we introduce a Supervised Contrastive knowledge distiLlation for class Incremental Fault Diagnosis (SCLIFD) framework proposing supervised contrastive knowledge distillation for improved representation learning capability and less forgetting, a novel prioritized exemplar selection method for sample replay to alleviate catastrophic forgetting, and the Random Forest Classifier to address the class imbalance. Extensive experimentation on simulated and real-world industrial datasets across various imbalance ratios demonstrates the superiority of SCLIFD over existing approaches. Our code can be found at https://github.com/Zhang-Henry/SCLIFD_TII.
☆ Merging Models on the Fly Without Retraining: A Sequential Approach to Scalable Continual Model Merging
Deep model merging represents an emerging research direction that combines multiple fine-tuned models to harness their specialized capabilities across different tasks and domains. Current model merging techniques focus on merging all available models simultaneously, with weight interpolation-based methods being the predominant approaches. However, these conventional approaches are not well-suited for scenarios where models become available sequentially, and they often suffer from high memory requirements and potential interference between tasks. In this study, we propose a training-free projection-based continual merging method that processes models sequentially through orthogonal projections of weight matrices and adaptive scaling mechanisms. Our method operates by projecting new parameter updates onto subspaces orthogonal to existing merged parameter updates while using an adaptive scaling mechanism to maintain stable parameter distances, enabling efficient sequential integration of task-specific knowledge. Our approach maintains constant memory complexity to the number of models, minimizes interference between tasks through orthogonal projections, and retains the performance of previously merged models through adaptive task vector scaling. Extensive experiments on CLIP-ViT models demonstrate that our method achieves a 5-8% average accuracy improvement while maintaining robust performance in different task orderings.
☆ Multi-task deep-learning for sleep event detection and stage classification
Polysomnographic sleep analysis is the standard clinical method to accurately diagnose and treat sleep disorders. It is an intricate process which involves the manual identification, classification, and location of multiple sleep event patterns. This is complex, for which identification of different types of events involves focusing on different subsets of signals, resulting on an iterative time-consuming process entailing several visual analysis passes. In this paper we propose a multi-task deep-learning approach for the simultaneous detection of sleep events and hypnogram construction in one single pass. Taking as reference state-of-the-art methodology for object-detection in the field of Computer Vision, we reformulate the problem for the analysis of multi-variate time sequences, and more specifically for pattern detection in the sleep analysis scenario. We investigate the performance of the resulting method in identifying different assembly combinations of EEG arousals, respiratory events (apneas and hypopneas) and sleep stages, also considering different input signal montage configurations. Furthermore, we evaluate our approach using two independent datasets, assessing true-generalization effects involving local and external validation scenarios. Based on our results, we analyze and discuss our method's capabilities and its potential wide-range applicability across different settings and datasets.
☆ PIER: A Novel Metric for Evaluating What Matters in Code-Switching ICASSP 2025
Code-switching, the alternation of languages within a single discourse, presents a significant challenge for Automatic Speech Recognition. Despite the unique nature of the task, performance is commonly measured with established metrics such as Word-Error-Rate (WER). However, in this paper, we question whether these general metrics accurately assess performance on code-switching. Specifically, using both Connectionist-Temporal-Classification and Encoder-Decoder models, we show fine-tuning on non-code-switched data from both matrix and embedded language improves classical metrics on code-switching test sets, although actual code-switched words worsen (as expected). Therefore, we propose Point-of-Interest Error Rate (PIER), a variant of WER that focuses only on specific words of interest. We instantiate PIER on code-switched utterances and show that this more accurately describes the code-switching performance, showing huge room for improvement in future work. This focused evaluation allows for a more precise assessment of model performance, particularly in challenging aspects such as inter-word and intra-word code-switching.
comment: Accepted at ICASSP 2025
☆ Multimodal Marvels of Deep Learning in Medical Diagnosis: A Comprehensive Review of COVID-19 Detection
This study presents a comprehensive review of the potential of multimodal deep learning (DL) in medical diagnosis, using COVID-19 as a case example. Motivated by the success of artificial intelligence applications during the COVID-19 pandemic, this research aims to uncover the capabilities of DL in disease screening, prediction, and classification, and to derive insights that enhance the resilience, sustainability, and inclusiveness of science, technology, and innovation systems. Adopting a systematic approach, we investigate the fundamental methodologies, data sources, preprocessing steps, and challenges encountered in various studies and implementations. We explore the architecture of deep learning models, emphasising their data-specific structures and underlying algorithms. Subsequently, we compare different deep learning strategies utilised in COVID-19 analysis, evaluating them based on methodology, data, performance, and prerequisites for future research. By examining diverse data types and diagnostic modalities, this research contributes to scientific understanding and knowledge of the multimodal application of DL and its effectiveness in diagnosis. We have implemented and analysed 11 deep learning models using COVID-19 image, text, and speech (ie, cough) data. Our analysis revealed that the MobileNet model achieved the highest accuracy of 99.97% for COVID-19 image data and 93.73% for speech data (i.e., cough). However, the BiGRU model demonstrated superior performance in COVID-19 text classification with an accuracy of 99.89%. The broader implications of this research suggest potential benefits for other domains and disciplines that could leverage deep learning techniques for image, text, and speech analysis.
comment: 43 pages
☆ MonoSOWA: Scalable monocular 3D Object detector Without human Annotations
Detecting the three-dimensional position and orientation of objects using a single RGB camera is a foundational task in computer vision with many important applications. Traditionally, 3D object detection methods are trained in a fully-supervised setup, requiring vast amounts of human annotations, which are laborious, costly, and do not scale well with the ever-increasing amounts of data being captured. In this paper, we present the first method to train 3D object detectors for monocular RGB cameras without domain-specific human annotations, thus making orders of magnitude more data available for training. Thanks to newly proposed Canonical Object Space, the method can not only exploit data across a variety of datasets and camera setups to train a single 3D detector, but unlike previous work it also works out of the box in previously unseen camera setups. All this is crucial for practical applications, where the data and cameras are extremely heterogeneous. The method is evaluated on two standard autonomous driving datasets, where it outperforms previous works, which, unlike our method, still rely on 2D human annotations.
☆ Utilizing AI Language Models to Identify Prognostic Factors for Coronary Artery Disease: A Study in Mashhad Residents
Abstract: Background: Understanding cardiovascular artery disease risk factors, the leading global cause of mortality, is crucial for influencing its etiology, prevalence, and treatment. This study aims to evaluate prognostic markers for coronary artery disease in Mashhad using Naive Bayes, REP Tree, J48, CART, and CHAID algorithms. Methods: Using data from the 2009 MASHAD STUDY, prognostic factors for coronary artery disease were determined with Naive Bayes, REP Tree, J48, CART, CHAID, and Random Forest algorithms using R 3.5.3 and WEKA 3.9.4. Model efficiency was compared by sensitivity, specificity, and accuracy. Cases were patients with coronary artery disease; each had three controls (totally 940). Results: Prognostic factors for coronary artery disease in Mashhad residents varied by algorithm. CHAID identified age, myocardial infarction history, and hypertension. CART included depression score and physical activity. REP added education level and anxiety score. NB included diabetes and family history. J48 highlighted father's heart disease and weight loss. CHAID had the highest accuracy (0.80). Conclusion: Key prognostic factors for coronary artery disease in CART and CHAID models include age, myocardial infarction history, hypertension, depression score, physical activity, and BMI. NB, REP Tree, and J48 identified numerous factors. CHAID had the highest accuracy, sensitivity, and specificity. CART offers simpler interpretation, aiding physician and paramedic model selection based on specific. Keywords: RF, Na\"ive Bayes, REP, J48 algorithms, Coronary Artery Disease (CAD).
☆ Predicting Air Temperature from Volumetric Urban Morphology with Machine Learning
In this study, we firstly introduce a method that converts CityGML data into voxels which works efficiently and fast in high resolution for large scale datasets such as cities but by sacrificing some building details to overcome the limitations of previous voxelization methodologies that have been computationally intensive and inefficient at transforming large-scale urban areas into voxel representations for high resolution. Those voxelized 3D city data from multiple cities and corresponding air temperature data are used to develop a machine learning model. Before the model training, Gaussian blurring is implemented on input data to consider spatial relationships, as a result the correlation rate between air temperature and volumetric building morphology is also increased after the Gaussian blurring. After the model training, the prediction results are not just evaluated with Mean Square Error (MSE) but some image similarity metrics such as Structural Similarity Index Measure (SSIM) and Learned Perceptual Image Patch Similarity (LPIPS) that are able to detect and consider spatial relations during the evaluation process. This trained model is capable of predicting the spatial distribution of air temperature by using building volume information of corresponding pixel as input. By doing so, this research aims to assist urban planners in incorporating environmental parameters into their planning strategies, thereby facilitating more sustainable and inhabitable urban environments.
comment: 30 pages, 8 figures, 2 tables
☆ Pruning for Sparse Diffusion Models based on Gradient Flow ICASSP2025
Diffusion Models (DMs) have impressive capabilities among generation models, but are limited to slower inference speeds and higher computational costs. Previous works utilize one-shot structure pruning to derive lightweight DMs from pre-trained ones, but this approach often leads to a significant drop in generation quality and may result in the removal of crucial weights. Thus we propose a iterative pruning method based on gradient flow, including the gradient flow pruning process and the gradient flow pruning criterion. We employ a progressive soft pruning strategy to maintain the continuity of the mask matrix and guide it along the gradient flow of the energy function based on the pruning criterion in sparse space, thereby avoiding the sudden information loss typically caused by one-shot pruning. Gradient-flow based criterion prune parameters whose removal increases the gradient norm of loss function and can enable fast convergence for a pruned model in iterative pruning stage. Our extensive experiments on widely used datasets demonstrate that our method achieves superior performance in efficiency and consistency with pre-trained models.
comment: 5 pages, 1 figure, accepted by ICASSP2025
☆ Teaching Wav2Vec2 the Language of the Brain ICASSP 2025
The decoding of continuously spoken speech from neuronal activity has the potential to become an important clinical solution for paralyzed patients. Deep Learning Brain Computer Interfaces (BCIs) have recently successfully mapped neuronal activity to text contents in subjects who attempted to formulate speech. However, only small BCI datasets are available. In contrast, labeled data and pre-trained models for the closely related task of speech recognition from audio are widely available. One such model is Wav2Vec2 which has been trained in a self-supervised fashion to create meaningful representations of speech audio data. In this study, we show that patterns learned by Wav2Vec2 are transferable to brain data. Specifically, we replace its audio feature extractor with an untrained Brain Feature Extractor (BFE) model. We then execute full fine-tuning with pre-trained weights for Wav2Vec2, training ''from scratch'' without pre-trained weights as well as freezing a pre-trained Wav2Vec2 and training only the BFE each for 45 different BFE architectures. Across these experiments, the best run is from full fine-tuning with pre-trained weights, achieving a Character Error Rate (CER) of 18.54\%, outperforming the best training from scratch run by 20.46\% and that of frozen Wav2Vec2 training by 15.92\% percentage points. These results indicate that knowledge transfer from audio speech recognition to brain decoding is possible and significantly improves brain decoding performance for the same architectures. Related source code is available at https://github.com/tfiedlerdev/Wav2Vec2ForBrain.
comment: Paper was submitted to ICASSP 2025 but marginally rejected
☆ Solving the unsolvable: Translating case law in Hong Kong
This paper addresses the challenges translating case law under Hong Kong's bilingual legal system. It highlights the initial success of translating all written statutes into Chinese before the 1997 handover, a task mandated by the Basic Law. The effort involved significant collaboration among legal, linguistic, and translation experts, resulting in a comprehensive and culturally appropriate bilingual legal system. However, translating case law remains a significant challenge due to the sheer volume and continuous growth of judicial decisions. The paper critiques the governments and judiciarys sporadic and uncoordinated efforts to translate case law, contrasting it with the thorough approach previously taken for statute translation. Although the government acknowledges the importance of legal bilingualism, it lacks a sustainable strategy for translating case law. The Judiciarys position that translating all judgments is unnecessary, unrealistic, and not cost-effectiveis analyzed and critiqued for its impact on legal transparency and public trust. A proposed solution involves leveraging machine translation technology through a human-machine interactive translation platform, which undergoes two major transitions. Initially based on a neural model, the platform transitions to using a large language model for improved translation accuracy. Furthermore, it evolves from a single-agent system to a multi-agent system, incorporating Translator, Annotator, and Proofreader agents. This multi-agent approach, supported by a grant, aims to facilitate efficient, high-quality translation of judicial judgments by integrating advanced artificial intelligence and continuous feedback mechanisms, thus better meeting the needs of a bilingual legal system.
☆ ADAGE: A generic two-layer framework for adaptive agent based modelling AAMAS
Agent-based models (ABMs) are valuable for modelling complex, potentially out-of-equilibria scenarios. However, ABMs have long suffered from the Lucas critique, stating that agent behaviour should adapt to environmental changes. Furthermore, the environment itself often adapts to these behavioural changes, creating a complex bi-level adaptation problem. Recent progress integrating multi-agent reinforcement learning into ABMs introduces adaptive agent behaviour, beginning to address the first part of this critique, however, the approaches are still relatively ad hoc, lacking a general formulation, and furthermore, do not tackle the second aspect of simultaneously adapting environmental level characteristics in addition to the agent behaviours. In this work, we develop a generic two-layer framework for ADaptive AGEnt based modelling (ADAGE) for addressing these problems. This framework formalises the bi-level problem as a Stackelberg game with conditional behavioural policies, providing a consolidated framework for adaptive agent-based modelling based on solving a coupled set of non-linear equations. We demonstrate how this generic approach encapsulates several common (previously viewed as distinct) ABM tasks, such as policy design, calibration, scenario generation, and robust behavioural learning under one unified framework. We provide example simulations on multiple complex economic and financial environments, showing the strength of the novel framework under these canonical settings, addressing long-standing critiques of traditional ABMs.
comment: Accepted at the 2025 International Conference on Autonomous Agents and Multiagent Systems (AAMAS)
☆ Dynamic Neural Style Transfer for Artistic Image Generation using VGG19
Throughout history, humans have created remarkable works of art, but artificial intelligence has only recently started to make strides in generating visually compelling art. Breakthroughs in the past few years have focused on using convolutional neural networks (CNNs) to separate and manipulate the content and style of images, applying texture synthesis techniques. Nevertheless, a number of current techniques continue to encounter obstacles, including lengthy processing times, restricted choices of style images, and the inability to modify the weight ratio of styles. We proposed a neural style transfer system that can add various artistic styles to a desired image to address these constraints allowing flexible adjustments to style weight ratios and reducing processing time. The system uses the VGG19 model for feature extraction, ensuring high-quality, flexible stylization without compromising content integrity.
☆ FASP: Fast and Accurate Structured Pruning of Large Language Models
The rapid increase in the size of large language models (LLMs) has significantly escalated their computational and memory demands, posing challenges for efficient deployment, especially on resource-constrained devices. Structured pruning has emerged as an effective model compression method that can reduce these demands while preserving performance. In this paper, we introduce FASP (Fast and Accurate Structured Pruning), a novel structured pruning framework for LLMs that emphasizes both speed and accuracy. FASP employs a distinctive pruning structure that interlinks sequential layers, allowing for the removal of columns in one layer while simultaneously eliminating corresponding rows in the preceding layer without incurring additional performance loss. The pruning metric, inspired by Wanda, is computationally efficient and effectively selects components to prune. Additionally, we propose a restoration mechanism that enhances model fidelity by adjusting the remaining weights post-pruning. We evaluate FASP on the OPT and LLaMA model families, demonstrating superior performance in terms of perplexity and accuracy on downstream tasks compared to state-of-the-art methods. Our approach achieves significant speed-ups, pruning models such as OPT-125M in 17 seconds and LLaMA-30B in 15 minutes on a single NVIDIA RTX 4090 GPU, making it a highly practical solution for optimizing LLMs.
☆ MoE$^2$: Optimizing Collaborative Inference for Edge Large Language Models
Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of natural language processing tasks. Exploiting the heterogeneous capabilities of edge LLMs is crucial for diverse emerging applications, as it enables greater cost-effectiveness and reduced latency. In this work, we introduce \textit{Mixture-of-Edge-Experts (MoE$^2$)}, a novel collaborative inference framework for edge LLMs. We formulate the joint gating and expert selection problem to optimize inference performance under energy and latency constraints. Unlike conventional MoE problems, LLM expert selection is significantly more challenging due to the combinatorial nature and the heterogeneity of edge LLMs across various attributes. To this end, we propose a two-level expert selection mechanism through which we uncover an optimality-preserving property of gating parameters across expert selections. This property enables the decomposition of the training and selection processes, significantly reducing complexity. Furthermore, we leverage the objective's monotonicity and design a discrete monotonic optimization algorithm for optimal expert selection. We implement edge servers with NVIDIA Jetson AGX Orins and NVIDIA RTX 4090 GPUs, and perform extensive experiments. Our results validate that performance improvements of various LLM models and show that our MoE$^2$ method can achieve optimal trade-offs among different delay and energy budgets, and outperforms baselines under various system resource constraints.
comment: Submitted to IEEE/ACM Transactions on Networking
☆ PISCO: Self-Supervised k-Space Regularization for Improved Neural Implicit k-Space Representations of Dynamic MRI
Neural implicit k-space representations (NIK) have shown promising results for dynamic magnetic resonance imaging (MRI) at high temporal resolutions. Yet, reducing acquisition time, and thereby available training data, results in severe performance drops due to overfitting. To address this, we introduce a novel self-supervised k-space loss function $\mathcal{L}_\mathrm{PISCO}$, applicable for regularization of NIK-based reconstructions. The proposed loss function is based on the concept of parallel imaging-inspired self-consistency (PISCO), enforcing a consistent global k-space neighborhood relationship without requiring additional data. Quantitative and qualitative evaluations on static and dynamic MR reconstructions show that integrating PISCO significantly improves NIK representations. Particularly for high acceleration factors (R$\geq$54), NIK with PISCO achieves superior spatio-temporal reconstruction quality compared to state-of-the-art methods. Furthermore, an extensive analysis of the loss assumptions and stability shows PISCO's potential as versatile self-supervised k-space loss function for further applications and architectures. Code is available at: https://github.com/compai-lab/2025-pisco-spieker
☆ Fast Searching of Extreme Operating Conditions for Relay Protection Setting Calculation Based on Graph Neural Network and Reinforcement Learning
Searching for the Extreme Operating Conditions (EOCs) is one of the core problems of power system relay protection setting calculation. The current methods based on brute-force search, heuristic algorithms, and mathematical programming can hardly meet the requirements of today's power systems in terms of computation speed due to the drastic changes in operating conditions induced by renewables and power electronics. This paper proposes an EOC fast search method, named Graph Dueling Double Deep Q Network (Graph D3QN), which combines graph neural network and deep reinforcement learning to address this challenge. First, the EOC search problem is modeled as a Markov decision process, where the information of the underlying power system is extracted using graph neural networks, so that the EOC of the system can be found via deep reinforcement learning. Then, a two-stage Guided Learning and Free Exploration (GLFE) training framework is constructed to accelerate the convergence speed of reinforcement learning. Finally, the proposed Graph D3QN method is validated through case studies of searching maximum fault current for relay protection setting calculation on the IEEE 39-bus and 118-bus systems. The experimental results demonstrate that Graph D3QN can reduce the computation time by 10 to 1000 times while guaranteeing the accuracy of the selected EOCs.
comment: 10 pages, 9 figures
☆ ELM-DeepONets: Backpropagation-Free Training of Deep Operator Networks via Extreme Learning Machines
Deep Operator Networks (DeepONets) are among the most prominent frameworks for operator learning, grounded in the universal approximation theorem for operators. However, training DeepONets typically requires significant computational resources. To address this limitation, we propose ELM-DeepONets, an Extreme Learning Machine (ELM) framework for DeepONets that leverages the backpropagation-free nature of ELM. By reformulating DeepONet training as a least-squares problem for newly introduced parameters, the ELM-DeepONet approach significantly reduces training complexity. Validation on benchmark problems, including nonlinear ODEs and PDEs, demonstrates that the proposed method not only achieves superior accuracy but also drastically reduces computational costs. This work offers a scalable and efficient alternative for operator learning in scientific computing.
☆ Quantum-Enhanced Transformers for Robust Acoustic Scene Classification in IoT Environments
The proliferation of Internet of Things (IoT) devices equipped with acoustic sensors necessitates robust acoustic scene classification (ASC) capabilities, even in noisy and data-limited environments. Traditional machine learning methods often struggle to generalize effectively under such conditions. To address this, we introduce Q-ASC, a novel Quantum-Inspired Acoustic Scene Classifier that leverages the power of quantum-inspired transformers. By integrating quantum concepts like superposition and entanglement, Q-ASC achieves superior feature learning and enhanced noise resilience compared to classical models. Furthermore, we introduce a Quantum Variational Autoencoder (QVAE) based data augmentation technique to mitigate the challenge of limited labeled data in IoT deployments. Extensive evaluations on the Tampere University of Technology (TUT) Acoustic Scenes 2016 benchmark dataset demonstrate that Q-ASC achieves remarkable accuracy between 68.3% and 88.5% under challenging conditions, outperforming state-of-the-art methods by over 5% in the best case. This research paves the way for deploying intelligent acoustic sensing in IoT networks, with potential applications in smart homes, industrial monitoring, and environmental surveillance, even in adverse acoustic environments.
comment: 5 pages, 4 figures
☆ PAL: Prompting Analytic Learning with Missing Modality for Multi-Modal Class-Incremental Learning
Multi-modal class-incremental learning (MMCIL) seeks to leverage multi-modal data, such as audio-visual and image-text pairs, thereby enabling models to learn continuously across a sequence of tasks while mitigating forgetting. While existing studies primarily focus on the integration and utilization of multi-modal information for MMCIL, a critical challenge remains: the issue of missing modalities during incremental learning phases. This oversight can exacerbate severe forgetting and significantly impair model performance. To bridge this gap, we propose PAL, a novel exemplar-free framework tailored to MMCIL under missing-modality scenarios. Concretely, we devise modality-specific prompts to compensate for missing information, facilitating the model to maintain a holistic representation of the data. On this foundation, we reformulate the MMCIL problem into a Recursive Least-Squares task, delivering an analytical linear solution. Building upon these, PAL not only alleviates the inherent under-fitting limitation in analytic learning but also preserves the holistic representation of missing-modality data, achieving superior performance with less forgetting across various multi-modal incremental scenarios. Extensive experiments demonstrate that PAL significantly outperforms competitive methods across various datasets, including UPMC-Food101 and N24News, showcasing its robustness towards modality absence and its anti-forgetting ability to maintain high incremental accuracy.
☆ Rational Tuning of LLM Cascades via Probabilistic Modeling
Understanding the reliability of large language models (LLMs) has recently garnered significant attention. Given LLMs' propensity to hallucinate, as well as their high sensitivity to prompt design, it is already challenging to predict the performance of an individual LLM. However, the problem becomes more complex for compound LLM systems such as cascades, where in addition to each model's standalone performance, we must understand how the error rates of different models interact. In this paper, we present a probabilistic model for the joint performance distribution of a sequence of LLMs, which enables a framework for rationally tuning the confidence thresholds of a LLM cascade using continuous optimization. Compared to selecting confidence thresholds using grid search, our parametric Markov-copula model significantly improves runtime scaling with respect to the length of the cascade and the desired resolution of the cost-error curve, turning them from intractable into low-order polynomial. In addition, the optimal thresholds computed using our continuous optimization-based algorithm increasingly outperform those found via grid search as cascade length grows, improving the area under the cost-error curve by 1.9% on average for cascades consisting of at least three models. Overall, our Markov-copula model provides a rational basis for tuning LLM cascade performance and points to the potential of probabilistic methods in analyzing LLM systems.
☆ Estimating shared subspace with AJIVE: the power and limitation of multiple data matrices
Integrative data analysis often requires disentangling joint and individual variations across multiple datasets, a challenge commonly addressed by the Joint and Individual Variation Explained (JIVE) model. While numerous methods have been developed to estimate the shared subspace under JIVE, the theoretical understanding of their performance remains limited, particularly in the context of multiple matrices and varying levels of subspace misalignment. This paper bridges this gap by providing a systematic analysis of shared subspace estimation in multi-matrix settings. We focus on the Angle-based Joint and Individual Variation Explained (AJIVE) method, a two-stage spectral approach, and establish new performance guarantees that uncover its strengths and limitations. Specifically, we show that in high signal-to-noise ratio (SNR) regimes, AJIVE's estimation error decreases with the number of matrices, demonstrating the power of multi-matrix integration. Conversely, in low-SNR settings, AJIVE exhibits a non-diminishing error, highlighting fundamental limitations. To complement these results, we derive minimax lower bounds, showing that AJIVE achieves optimal rates in high-SNR regimes. Furthermore, we analyze an oracle-aided spectral estimator to demonstrate that the non-diminishing error in low-SNR scenarios is a fundamental barrier. Extensive numerical experiments corroborate our theoretical findings, providing insights into the interplay between SNR, matrix count, and subspace misalignment.
☆ Identifying Information from Observations with Uncertainty and Novelty
A machine learning tasks from observations must encounter and process uncertainty and novelty, especially when it is expected to maintain performance when observing new information and to choose the best fitting hypothesis to the currently observed information. In this context, some key questions arise: what is information, how much information did the observations provide, how much information is required to identify the data-generating process, how many observations remain to get that information, and how does a predictor determine that it has observed novel information? This paper strengthens existing answers to these questions by formalizing the notion of "identifiable information" that arises from the language used to express the relationship between distinct states. Model identifiability and sample complexity are defined via computation of an indicator function over a set of hypotheses. Their properties and asymptotic statistics are described for data-generating processes ranging from deterministic processes to ergodic stationary stochastic processes. This connects the notion of identifying information in finite steps with asymptotic statistics and PAC-learning. The indicator function's computation naturally formalizes novel information and its identification from observations with respect to a hypothesis set. We also proved that computable PAC-Bayes learners' sample complexity distribution is determined by its moments in terms of the the prior probability distribution over a fixed finite hypothesis set.
comment: 43 pages, 1 figure, 1 table, and 2 inline algorithms. Submitted to JMLR Jan. 6, 2025
☆ On Learning Informative Trajectory Embeddings for Imitation, Classification and Regression AAMAS 2025
In real-world sequential decision making tasks like autonomous driving, robotics, and healthcare, learning from observed state-action trajectories is critical for tasks like imitation, classification, and clustering. For example, self-driving cars must replicate human driving behaviors, while robots and healthcare systems benefit from modeling decision sequences, whether or not they come from expert data. Existing trajectory encoding methods often focus on specific tasks or rely on reward signals, limiting their ability to generalize across domains and tasks. Inspired by the success of embedding models like CLIP and BERT in static domains, we propose a novel method for embedding state-action trajectories into a latent space that captures the skills and competencies in the dynamic underlying decision-making processes. This method operates without the need for reward labels, enabling better generalization across diverse domains and tasks. Our contributions are threefold: (1) We introduce a trajectory embedding approach that captures multiple abilities from state-action data. (2) The learned embeddings exhibit strong representational power across downstream tasks, including imitation, classification, clustering, and regression. (3) The embeddings demonstrate unique properties, such as controlling agent behaviors in IQ-Learn and an additive structure in the latent space. Experimental results confirm that our method outperforms traditional approaches, offering more flexible and powerful trajectory representations for various applications. Our code is available at https://github.com/Erasmo1015/vte.
comment: AAMAS 2025
☆ Cooperative Decentralized Backdoor Attacks on Vertical Federated Learning
Federated learning (FL) is vulnerable to backdoor attacks, where adversaries alter model behavior on target classification labels by embedding triggers into data samples. While these attacks have received considerable attention in horizontal FL, they are less understood for vertical FL (VFL), where devices hold different features of the samples, and only the server holds the labels. In this work, we propose a novel backdoor attack on VFL which (i) does not rely on gradient information from the server and (ii) considers potential collusion among multiple adversaries for sample selection and trigger embedding. Our label inference model augments variational autoencoders with metric learning, which adversaries can train locally. A consensus process over the adversary graph topology determines which datapoints to poison. We further propose methods for trigger splitting across the adversaries, with an intensity-based implantation scheme skewing the server towards the trigger. Our convergence analysis reveals the impact of backdoor perturbations on VFL indicated by a stationarity gap for the trained model, which we verify empirically as well. We conduct experiments comparing our attack with recent backdoor VFL approaches, finding that ours obtains significantly higher success rates for the same main task performance despite not using server information. Additionally, our results verify the impact of collusion on attack performance.
comment: This paper is currently under review in the IEEE/ACM Transactions on Networking Special Issue on AI and Networking
☆ Finding the Trigger: Causal Abductive Reasoning on Video Events
This paper introduces a new problem, Causal Abductive Reasoning on Video Events (CARVE), which involves identifying causal relationships between events in a video and generating hypotheses about causal chains that account for the occurrence of a target event. To facilitate research in this direction, we create two new benchmark datasets with both synthetic and realistic videos, accompanied by trigger-target labels generated through a novel counterfactual synthesis approach. To explore the challenge of solving CARVE, we present a Causal Event Relation Network (CERN) that examines the relationships between video events in temporal and semantic spaces to efficiently determine the root-cause trigger events. Through extensive experiments, we demonstrate the critical roles of event relational representation learning and interaction modeling in solving video causal reasoning challenges. The introduction of the CARVE task, along with the accompanying datasets and the CERN framework, will advance future research on video causal reasoning and significantly facilitate various applications, including video surveillance, root-cause analysis and movie content management.
☆ Physics-informed deep learning for infectious disease forecasting
Accurate forecasting of contagious illnesses has become increasingly important to public health policymaking, and better prediction could prevent the loss of millions of lives. To better prepare for future pandemics, it is essential to improve forecasting methods and capabilities. In this work, we propose a new infectious disease forecasting model based on physics-informed neural networks (PINNs), an emerging area of scientific machine learning. The proposed PINN model incorporates dynamical systems representations of disease transmission into the loss function, thereby assimilating epidemiological theory and data using neural networks (NNs). Our approach is designed to prevent model overfitting, which often occurs when training deep learning models with observation data alone. In addition, we employ an additional sub-network to account for mobility, vaccination, and other covariates that influence the transmission rate, a key parameter in the compartment model. To demonstrate the capability of the proposed model, we examine the performance of the model using state-level COVID-19 data in California. Our simulation results show that predictions of PINN model on the number of cases, deaths, and hospitalizations are consistent with existing benchmarks. In particular, the PINN model outperforms the basic NN model and naive baseline forecast. We also show that the performance of the PINN model is comparable to a sophisticated Gaussian infection state space with time dependence (GISST) forecasting model that integrates the compartment model with a data observation model and a regression model for inferring parameters in the compartment model. Nonetheless, the PINN model offers a simpler structure and is easier to implement. Our results show that the proposed forecaster could potentially serve as a new computational tool to enhance the current capacity of infectious disease forecasting.
☆ Free-Knots Kolmogorov-Arnold Network: On the Analysis of Spline Knots and Advancing Stability
Kolmogorov-Arnold Neural Networks (KANs) have gained significant attention in the machine learning community. However, their implementation often suffers from poor training stability and heavy trainable parameter. Furthermore, there is limited understanding of the behavior of the learned activation functions derived from B-splines. In this work, we analyze the behavior of KANs through the lens of spline knots and derive the lower and upper bound for the number of knots in B-spline-based KANs. To address existing limitations, we propose a novel Free Knots KAN that enhances the performance of the original KAN while reducing the number of trainable parameters to match the trainable parameter scale of standard Multi-Layer Perceptrons (MLPs). Additionally, we introduce new a training strategy to ensure $C^2$ continuity of the learnable spline, resulting in smoother activation compared to the original KAN and improve the training stability by range expansion. The proposed method is comprehensively evaluated on 8 datasets spanning various domains, including image, text, time series, multimodal, and function approximation tasks. The promising results demonstrates the feasibility of KAN-based network and the effectiveness of proposed method.
☆ Large Language Model is Secretly a Protein Sequence Optimizer
We consider the protein sequence engineering problem, which aims to find protein sequences with high fitness levels, starting from a given wild-type sequence. Directed evolution has been a dominating paradigm in this field which has an iterative process to generate variants and select via experimental feedback. We demonstrate large language models (LLMs), despite being trained on massive texts, are secretly protein sequence optimizers. With a directed evolutionary method, LLM can perform protein engineering through Pareto and experiment-budget constrained optimization, demonstrating success on both synthetic and experimental fitness landscapes.
comment: Preprint
☆ On the convergence of noisy Bayesian Optimization with Expected Improvement
Expected improvement (EI) is one of the most widely-used acquisition functions in Bayesian optimization (BO). Despite its proven success in applications for decades, important open questions remain on the theoretical convergence behaviors and rates for EI. In this paper, we contribute to the convergence theories of EI in three novel and critical area. First, we consider objective functions that are under the Gaussian process (GP) prior assumption, whereas existing works mostly focus on functions in the reproducing kernel Hilbert space (RKHS). Second, we establish the first asymptotic error bound and its corresponding rate for GP-EI with noisy observations under the GP prior assumption. Third, by investigating the exploration and exploitation of the non-convex EI function, we prove improved error bounds for both the noise-free and noisy cases. The improved noiseless bound is extended to the RKHS assumption as well.
☆ Clone-Robust AI Alignment
A key challenge in training Large Language Models (LLMs) is properly aligning them with human preferences. Reinforcement Learning with Human Feedback (RLHF) uses pairwise comparisons from human annotators to train reward functions and has emerged as a popular alignment method. However, input datasets in RLHF are not necessarily balanced in the types of questions and answers that are included. Therefore, we want RLHF algorithms to perform well even when the set of alternatives is not uniformly distributed. Drawing on insights from social choice theory, we introduce robustness to approximate clones, a desirable property of RLHF algorithms which requires that adding near-duplicate alternatives does not significantly change the learned reward function. We first demonstrate that the standard RLHF algorithm based on regularized maximum likelihood estimation (MLE) fails to satisfy this property. We then propose the weighted MLE, a new RLHF algorithm that modifies the standard regularized MLE by weighting alternatives based on their similarity to other alternatives. This new algorithm guarantees robustness to approximate clones while preserving desirable theoretical properties.
☆ Task Vectors in In-Context Learning: Emergence, Formation, and Benefit
In-context learning is a remarkable capability of transformers, referring to their ability to adapt to specific tasks based on a short history or context. Previous research has found that task-specific information is locally encoded within models, though their emergence and functionality remain unclear due to opaque pre-training processes. In this work, we investigate the formation of task vectors in a controlled setting, using models trained from scratch on synthetic datasets. Our findings confirm that task vectors naturally emerge under certain conditions, but the tasks may be relatively weakly and/or non-locally encoded within the model. To promote strong task vectors encoded at a prescribed location within the model, we propose an auxiliary training mechanism based on a task vector prompting loss (TVP-loss). This method eliminates the need to search for task-correlated encodings within the trained model and demonstrably improves robustness and generalization.
☆ Mono-Forward: Backpropagation-Free Algorithm for Efficient Neural Network Training Harnessing Local Errors
Backpropagation is the standard method for achieving state-of-the-art accuracy in neural network training, but it often imposes high memory costs and lacks biological plausibility. In this paper, we introduce the Mono-Forward algorithm, a purely local layerwise learning method inspired by Hinton's Forward-Forward framework. Unlike backpropagation, Mono-Forward optimizes each layer solely with locally available information, eliminating the reliance on global error signals. We evaluated Mono-Forward on multi-layer perceptrons and convolutional neural networks across multiple benchmarks, including MNIST, Fashion-MNIST, CIFAR-10, and CIFAR-100. The test results show that Mono-Forward consistently matches or surpasses the accuracy of backpropagation across all tasks, with significantly reduced and more even memory usage, better parallelizability, and a comparable convergence rate.
comment: 12 pages
☆ Tessellated Linear Model for Age Prediction from Voice
Voice biometric tasks, such as age estimation require modeling the often complex relationship between voice features and the biometric variable. While deep learning models can handle such complexity, they typically require large amounts of accurately labeled data to perform well. Such data are often scarce for biometric tasks such as voice-based age prediction. On the other hand, simpler models like linear regression can work with smaller datasets but often fail to generalize to the underlying non-linear patterns present in the data. In this paper we propose the Tessellated Linear Model (TLM), a piecewise linear approach that combines the simplicity of linear models with the capacity of non-linear functions. TLM tessellates the feature space into convex regions and fits a linear model within each region. We optimize the tessellation and the linear models using a hierarchical greedy partitioning. We evaluated TLM on the TIMIT dataset on the task of age prediction from voice, where it outperformed state-of-the-art deep learning models.
☆ Foundations of Large Language Models
This is a book about large language models. As indicated by the title, it primarily focuses on foundational concepts rather than comprehensive coverage of all cutting-edge technologies. The book is structured into four main chapters, each exploring a key area: pre-training, generative models, prompting techniques, and alignment methods. It is intended for college students, professionals, and practitioners in natural language processing and related fields, and can serve as a reference for anyone interested in large language models.
☆ Leveraging Scale-aware Representations for improved Concept-Representation Alignment in ViTs
Vision Transformers (ViTs) are increasingly being adopted in various sensitive vision applications - like medical diagnosis, facial recognition, etc. To improve the interpretability of such models, many approaches attempt to forward-align them with carefully annotated abstract, human-understandable semantic entities - concepts. Concepts provide global rationales to the model predictions and can be quickly understood/intervened on by domain experts. Most current research focuses on designing model-agnostic, plug-and-play generic concept-based explainability modules that do not incorporate the inner workings of foundation models (e.g., inductive biases, scale invariance, etc.) during training. To alleviate this issue for ViTs, in this paper, we propose a novel Concept Representation Alignment Module (CRAM) which learns both scale and position-aware representations from multi-scale feature pyramids and patch representations respectively. CRAM further aligns these representations with concept annotations through an attention matrix. The proposed CRAM module improves the predictive performance of ViT architectures and also provides accurate and robust concept explanations as demonstrated on five datasets - including three widely used benchmarks (CUB, Pascal APY, Concept-MNIST) and 2 real-world datasets (AWA2, KITS).
☆ Adaptive Law-Based Transformation (ALT): A Lightweight Feature Representation for Time Series Classification
Time series classification (TSC) is fundamental in numerous domains, including finance, healthcare, and environmental monitoring. However, traditional TSC methods often struggle with the inherent complexity and variability of time series data. Building on our previous work with the linear law-based transformation (LLT) - which improved classification accuracy by transforming the feature space based on key data patterns - we introduce adaptive law-based transformation (ALT). ALT enhances LLT by incorporating variable-length shifted time windows, enabling it to capture distinguishing patterns of various lengths and thereby handle complex time series more effectively. By mapping features into a linearly separable space, ALT provides a fast, robust, and transparent solution that achieves state-of-the-art performance with only a few hyperparameters.
comment: 8 pages, 1 figure, 5 tables
☆ CLAP-S: Support Set Based Adaptation for Downstream Fiber-optic Acoustic Recognition ICASSP 2025
Contrastive Language-Audio Pretraining (CLAP) models have demonstrated unprecedented performance in various acoustic signal recognition tasks. Fiber-optic-based acoustic recognition is one of the most important downstream tasks and plays a significant role in environmental sensing. Adapting CLAP for fiber-optic acoustic recognition has become an active research area. As a non-conventional acoustic sensor, fiber-optic acoustic recognition presents a challenging, domain-specific, low-shot deployment environment with significant domain shifts due to unique frequency response and noise characteristics. To address these challenges, we propose a support-based adaptation method, CLAP-S, which linearly interpolates a CLAP Adapter with the Support Set, leveraging both implicit knowledge through fine-tuning and explicit knowledge retrieved from memory for cross-domain generalization. Experimental results show that our method delivers competitive performance on both laboratory-recorded fiber-optic ESC-50 datasets and a real-world fiber-optic gunshot-firework dataset. Our research also provides valuable insights for other downstream acoustic recognition tasks. The code and gunshot-firework dataset are available at https://github.com/Jingchensun/clap-s.
comment: Accepted to ICASSP 2025
☆ Geometry-Preserving Encoder/Decoder in Latent Generative Models
Generative modeling aims to generate new data samples that resemble a given dataset, with diffusion models recently becoming the most popular generative model. One of the main challenges of diffusion models is solving the problem in the input space, which tends to be very high-dimensional. Recently, solving diffusion models in the latent space through an encoder that maps from the data space to a lower-dimensional latent space has been considered to make the training process more efficient and has shown state-of-the-art results. The variational autoencoder (VAE) is the most commonly used encoder/decoder framework in this domain, known for its ability to learn latent representations and generate data samples. In this paper, we introduce a novel encoder/decoder framework with theoretical properties distinct from those of the VAE, specifically designed to preserve the geometric structure of the data distribution. We demonstrate the significant advantages of this geometry-preserving encoder in the training process of both the encoder and decoder. Additionally, we provide theoretical results proving convergence of the training process, including convergence guarantees for encoder training, and results showing faster convergence of decoder training when using the geometry-preserving encoder.
comment: 41 pages
☆ An LLM-Guided Tutoring System for Social Skills Training
Social skills training targets behaviors necessary for success in social interactions. However, traditional classroom training for such skills is often insufficient to teach effective communication -- one-to-one interaction in real-world scenarios is preferred to lecture-style information delivery. This paper introduces a framework that allows instructors to collaborate with large language models to dynamically design realistic scenarios for students to communicate. Our framework uses these scenarios to enable student rehearsal, provide immediate feedback, and visualize performance for both students and instructors. Unlike traditional intelligent tutoring systems, instructors can easily co-create scenarios with a large language model without technical skills. Additionally, the system generates new scenario branches in real time when existing options do not fit the student's response.
☆ From Explainability to Interpretability: Interpretable Policies in Reinforcement Learning Via Model Explanation AAAI
Deep reinforcement learning (RL) has shown remarkable success in complex domains, however, the inherent black box nature of deep neural network policies raises significant challenges in understanding and trusting the decision-making processes. While existing explainable RL methods provide local insights, they fail to deliver a global understanding of the model, particularly in high-stakes applications. To overcome this limitation, we propose a novel model-agnostic approach that bridges the gap between explainability and interpretability by leveraging Shapley values to transform complex deep RL policies into transparent representations. The proposed approach offers two key contributions: a novel approach employing Shapley values to policy interpretation beyond local explanations and a general framework applicable to off-policy and on-policy algorithms. We evaluate our approach with three existing deep RL algorithms and validate its performance in two classic control environments. The results demonstrate that our approach not only preserves the original models' performance but also generates more stable interpretable policies.
comment: Accepted to Deployable AI (DAI) Workshop at the Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI-25)
☆ Learning Noisy Halfspaces with a Margin: Massart is No Harder than Random NeurIPS 2024
We study the problem of PAC learning $\gamma$-margin halfspaces with Massart noise. We propose a simple proper learning algorithm, the Perspectron, that has sample complexity $\widetilde{O}((\epsilon\gamma)^{-2})$ and achieves classification error at most $\eta+\epsilon$ where $\eta$ is the Massart noise rate. Prior works [DGT19,CKMY20] came with worse sample complexity guarantees (in both $\epsilon$ and $\gamma$) or could only handle random classification noise [DDK+23,KIT+23] -- a much milder noise assumption. We also show that our results extend to the more challenging setting of learning generalized linear models with a known link function under Massart noise, achieving a similar sample complexity to the halfspace case. This significantly improves upon the prior state-of-the-art in this setting due to [CKMY20], who introduced this model.
comment: Appeared in NeurIPS 2024
☆ Coded Deep Learning: Framework and Algorithm
The success of deep learning (DL) is often achieved with large models and high complexity during both training and post-training inferences, hindering training in resource-limited settings. To alleviate these issues, this paper introduces a new framework dubbed ``coded deep learning'' (CDL), which integrates information-theoretic coding concepts into the inner workings of DL, to significantly compress model weights and activations, reduce computational complexity at both training and post-training inference stages, and enable efficient model/data parallelism. Specifically, within CDL, (i) we first propose a novel probabilistic method for quantizing both model weights and activations, and its soft differentiable variant which offers an analytic formula for gradient calculation during training; (ii) both the forward and backward passes during training are executed over quantized weights and activations, eliminating most floating-point operations and reducing training complexity; (iii) during training, both weights and activations are entropy constrained so that they are compressible in an information-theoretic sense throughout training, thus reducing communication costs in model/data parallelism; and (iv) the trained model in CDL is by default in a quantized format with compressible quantized weights, reducing post-training inference and storage complexity. Additionally, a variant of CDL, namely relaxed CDL (R-CDL), is presented to further improve the trade-off between validation accuracy and compression though requiring full precision in training with other advantageous features of CDL intact. Extensive empirical results show that CDL and R-CDL outperform the state-of-the-art algorithms in DNN compression in the literature.
☆ pFedWN: A Personalized Federated Learning Framework for D2D Wireless Networks with Heterogeneous Data
Traditional Federated Learning (FL) approaches often struggle with data heterogeneity across clients, leading to suboptimal model performance for individual clients. To address this issue, Personalized Federated Learning (PFL) emerges as a solution to the challenges posed by non-independent and identically distributed (non-IID) and unbalanced data across clients. Furthermore, in most existing decentralized machine learning works, a perfect communication channel is considered for model parameter transmission between clients and servers. However, decentralized PFL over wireless links introduces new challenges, such as resource allocation and interference management. To overcome these challenges, we formulate a joint optimization problem that incorporates the underlying device-to-device (D2D) wireless channel conditions into a server-free PFL approach. The proposed method, dubbed pFedWN, optimizes the learning performance for each client while accounting for the variability in D2D wireless channels. To tackle the formulated problem, we divide it into two sub-problems: PFL neighbor selection and PFL weight assignment. The PFL neighbor selection is addressed through channel-aware neighbor selection within unlicensed spectrum bands such as ISM bands. Next, to assign PFL weights, we utilize the Expectation-Maximization (EM) method to evaluate the similarity between clients' data and obtain optimal weight distribution among the chosen PFL neighbors. Empirical results show that pFedWN provides efficient and personalized learning performance with non-IID and unbalanced datasets. Furthermore, it outperforms the existing FL and PFL methods in terms of learning efficacy and robustness, particularly under dynamic and unpredictable wireless channel conditions.
comment: 16 pages, 9 figures, 3 tables, submitted to Transactions on Networking
☆ BN-Pool: a Bayesian Nonparametric Approach to Graph Pooling
We introduce BN-Pool, the first clustering-based pooling method for Graph Neural Networks (GNNs) that adaptively determines the number of supernodes in a coarsened graph. By leveraging a Bayesian non-parametric framework, BN-Pool employs a generative model capable of partitioning graph nodes into an unbounded number of clusters. During training, we learn the node-to-cluster assignments by combining the supervised loss of the downstream task with an unsupervised auxiliary term, which encourages the reconstruction of the original graph topology while penalizing unnecessary proliferation of clusters. This adaptive strategy allows BN-Pool to automatically discover an optimal coarsening level, offering enhanced flexibility and removing the need to specify sensitive pooling ratios. We show that BN-Pool achieves superior performance across diverse benchmarks.
☆ Enhancing Generalization in Chain of Thought Reasoning for Smaller Models
Chain-of-Thought (CoT) reasoning in smaller language models is a challenging natural language process problem yet highly desirable in many real-life applications. Existing CoT knowledge distillation methods often suffer from overly conservative memorization in smaller LLMs, leading to low generalization confidence. As fully preserving the CoT ability of teacher model is impossible, we hypothesize that adversarial CoT fine-tuning is crucial for developing smaller LLM with robust CoT generalization. To this end, we propose \textit{PRompt-Assisted Domain-Adversarial fine-tuning} (PRADA), a principled fine-tuning framework that integrates diverse CoT domains. Specifically, PRADA pioneers two CoT improvements in smaller LLM: (1) Recovering the domain-invariant feature insight which typically lost during distillation with domain adversarial fine-tuning; (2) Enhancing the domain adaptability of CoT prompt engineering by employing domain-adversarial approaches. We theoretically demonstrate the effectiveness of our approach and empirically show that it significantly outperforms the state of the arts in a wide range of tasks. Moreover, our empirical findings reveal that the smaller LLM, when leveraging PRADA, aligns closely with domain knowledge, thereby improving the explainability of our approach.
☆ Graph Neural Networks for Travel Distance Estimation and Route Recommendation Under Probabilistic Hazards
Estimating the shortest travel time and providing route recommendation between different locations in a city or region can quantitatively measure the conditions of the transportation network during or after extreme events. One common approach is to use Dijkstra's Algorithm, which produces the shortest path as well as the shortest distance. However, this option is computationally expensive when applied to large-scale networks. This paper proposes a novel fast framework based on graph neural networks (GNNs) which approximate the single-source shortest distance between pairs of locations, and predict the single-source shortest path subsequently. We conduct multiple experiments on synthetic graphs of different size to demonstrate the feasibility and computational efficiency of the proposed model. In real-world case studies, we also applied the proposed method of flood risk analysis of coastal urban areas to calculate delays in evacuation to public shelters during hurricanes. The results indicate the accuracy and computational efficiency of the GNN model, and its potential for effective implementation in emergency planning and management.
comment: 17 pages, 11 figures
☆ OmniThink: Expanding Knowledge Boundaries in Machine Writing through Thinking
Machine writing with large language models often relies on retrieval-augmented generation. However, these approaches remain confined within the boundaries of the model's predefined scope, limiting the generation of content with rich information. Specifically, vanilla-retrieved information tends to lack depth, utility, and suffers from redundancy, which negatively impacts the quality of generated articles, leading to shallow, repetitive, and unoriginal outputs. To address these issues, we propose OmniThink, a machine writing framework that emulates the human-like process of iterative expansion and reflection. The core idea behind OmniThink is to simulate the cognitive behavior of learners as they progressively deepen their knowledge of the topics. Experimental results demonstrate that OmniThink improves the knowledge density of generated articles without compromising metrics such as coherence and depth. Human evaluations and expert feedback further highlight the potential of OmniThink to address real-world challenges in the generation of long-form articles.
♻ ☆ Algorithmic Collective Action in Recommender Systems: Promoting Songs by Reordering Playlists NeurIPS 2024
We investigate algorithmic collective action in transformer-based recommender systems. Our use case is a music streaming platform where a collective of fans aims to promote the visibility of an underrepresented artist by strategically placing one of their songs in the existing playlists they control. We introduce two easily implementable strategies to select the position at which to insert the song with the goal to boost recommendations at test time. The strategies exploit statistical properties of the learner by targeting discontinuities in the recommendations, and leveraging the long-tail nature of song distributions. We evaluate the efficacy of our strategies using a publicly available recommender system model released by a major music streaming platform. Our findings reveal that through strategic placement even small collectives (controlling less than 0.01\% of the training data) can achieve up to $40\times$ more test time recommendations than an average song with the same number of training set occurrences. Focusing on the externalities of the strategy, we find that the recommendations of other songs are largely preserved, and the newly gained recommendations are distributed across various artists. Together, our findings demonstrate how carefully designed collective action strategies can be effective while not necessarily being adversarial.
comment: Published at NeurIPS 2024, camera-ready updates
♻ ☆ Using Machine Learning to Discover Parsimonious and Physically-Interpretable Representations of Catchment-Scale Rainfall-Runoff Dynamics
Despite the excellent real-world predictive performance of modern machine learning (ML) methods, many scientists remain hesitant to discard traditional physical-conceptual (PC) approaches due mainly to their relative interpretability, which contributes to credibility during decision-making. In this context, a currently underexplored aspect of ML is how to develop minimally-optimal representations that can facilitate better insight regarding system functioning. Regardless of how this is achieved, it is arguably true that parsimonious representations better support the advancement of scientific understanding. Our own view is that ML-based modeling of geoscientific systems should be based in the use of computational units that are fundamentally interpretable by design. This paper continues our exploration of how the strengths of ML can be exploited in the service of better understanding via scientific investigation. Here, we use the Mass Conserving Perceptron (MCP) as the fundamental computational unit in a generic network architecture consisting of nodes arranged in series and parallel to explore several generic and important issues related to the use of observational data for constructing input-state-output models of dynamical systems. In the context of lumped catchment modeling, we show that physical interpretability and excellent predictive performance can both be achieved using a relatively parsimonious distributed-state multiple-flow-path network with context-dependent gating and information sharing across the nodes, suggesting that MCP-based modeling can play a significant role in application of ML to geoscientific investigation.
comment: 74 Pages, 4 Tables, 13 Figures, 11 Tables and 11 Figures in Supplementary Materials
♻ ☆ Dynamics of Moral Behavior in Heterogeneous Populations of Learning Agents AAAI
Growing concerns about safety and alignment of AI systems highlight the importance of embedding moral capabilities in artificial agents: a promising solution is the use of learning from experience, i.e., Reinforcement Learning. In multi-agent (social) environments, complex population-level phenomena may emerge from interactions between individual learning agents. Many of the existing studies rely on simulated social dilemma environments to study the interactions of independent learning agents; however, they tend to ignore the moral heterogeneity that is likely to be present in societies of agents in practice. For example, at different points in time a single learning agent may face opponents who are consequentialist (i.e., focused on maximizing outcomes over time), norm-based (i.e., conforming to specific norms), or virtue-based (i.e., considering a combination of different virtues). The extent to which agents' co-development may be impacted by such moral heterogeneity in populations is not well understood. In this paper, we present a study of the learning dynamics of morally heterogeneous populations interacting in a social dilemma setting. Using an Iterated Prisoner's Dilemma environment with a partner selection mechanism, we investigate the extent to which the prevalence of diverse moral agents in populations affects individual agents' learning behaviors and emergent population-level outcomes. We observe several types of non-trivial interactions between pro-social and anti-social agents, and find that certain types of moral agents are able to steer selfish agents towards more cooperative behavior.
comment: Presented at AIES 2024 (7th AAAI/ACM Conference on AI, Ethics, and Society - San Jose, CA, USA) - see https://ojs.aaai.org/index.php/AIES/article/view/31736
♻ ☆ A Comparative Study on Multi-task Uncertainty Quantification in Semantic Segmentation and Monocular Depth Estimation
Deep neural networks excel in perception tasks such as semantic segmentation and monocular depth estimation, making them indispensable in safety-critical applications like autonomous driving and industrial inspection. However, they often suffer from overconfidence and poor explainability, especially for out-of-domain data. While uncertainty quantification has emerged as a promising solution to these challenges, multi-task settings have yet to be explored. In an effort to shed light on this, we evaluate Monte Carlo Dropout, Deep Sub-Ensembles, and Deep Ensembles for joint semantic segmentation and monocular depth estimation. Thereby, we reveal that Deep Ensembles stand out as the preferred choice, particularly in out-of-domain scenarios, and show the potential benefit of multi-task learning with regard to the uncertainty quality in comparison to solving both tasks separately. Additionally, we highlight the impact of employing different uncertainty thresholds to classify pixels as certain or uncertain, with the median uncertainty emerging as a robust default.
comment: This manuscript is an extended version of a previously published conference paper and is currently in review for a journal
♻ ☆ Aligning Brain Activity with Advanced Transformer Models: Exploring the Role of Punctuation in Semantic Processing
This research examines the congruence between neural activity and advanced transformer models, emphasizing the semantic significance of punctuation in text understanding. Utilizing an innovative approach originally proposed by Toneva and Wehbe, we evaluate four advanced transformer models RoBERTa, DistiliBERT, ALBERT, and ELECTRA against neural activity data. Our findings indicate that RoBERTa exhibits the closest alignment with neural activity, surpassing BERT in accuracy. Furthermore, we investigate the impact of punctuation removal on model performance and neural alignment, revealing that BERT's accuracy enhances in the absence of punctuation. This study contributes to the comprehension of how neural networks represent language and the influence of punctuation on semantic processing within the human brain.
♻ ☆ Flexible task abstractions emerge in linear networks with fast and bounded units
Animals survive in dynamic environments changing at arbitrary timescales, but such data distribution shifts are a challenge to neural networks. To adapt to change, neural systems may change a large number of parameters, which is a slow process involving forgetting past information. In contrast, animals leverage distribution changes to segment their stream of experience into tasks and associate them with internal task abstracts. Animals can then respond flexibly by selecting the appropriate task abstraction. However, how such flexible task abstractions may arise in neural systems remains unknown. Here, we analyze a linear gated network where the weights and gates are jointly optimized via gradient descent, but with neuron-like constraints on the gates including a faster timescale, nonnegativity, and bounded activity. We observe that the weights self-organize into modules specialized for tasks or sub-tasks encountered, while the gates layer forms unique representations that switch the appropriate weight modules (task abstractions). We analytically reduce the learning dynamics to an effective eigenspace, revealing a virtuous cycle: fast adapting gates drive weight specialization by protecting previous knowledge, while weight specialization in turn increases the update rate of the gating layer. Task switching in the gating layer accelerates as a function of curriculum block size and task training, mirroring key findings in cognitive neuroscience. We show that the discovered task abstractions support generalization through both task and subtask composition, and we extend our findings to a non-linear network switching between two tasks. Overall, our work offers a theory of cognitive flexibility in animals as arising from joint gradient descent on synaptic and neural gating in a neural network architecture.
♻ ☆ A Comprehensive Survey of Foundation Models in Medicine
Foundation models (FMs) are large-scale deep learning models trained on massive datasets, often using self-supervised learning techniques. These models serve as a versatile base for a wide range of downstream tasks, including those in medicine and healthcare. FMs have demonstrated remarkable success across multiple healthcare domains. However, existing surveys in this field do not comprehensively cover all areas where FMs have made significant strides. In this survey, we present a comprehensive review of FMs in medicine, focusing on their evolution, learning strategies, flagship models, applications, and associated challenges. We examine how prominent FMs, such as the BERT and GPT families, are transforming various aspects of healthcare, including clinical large language models, medical image analysis, and omics research. Additionally, we provide a detailed taxonomy of FM-enabled healthcare applications, spanning clinical natural language processing, medical computer vision, graph learning, and other biology- and omics- related tasks. Despite the transformative potentials of FMs, they also pose unique challenges. This survey delves into these challenges and highlights open research questions and lessons learned to guide researchers and practitioners. Our goal is to provide valuable insights into the capabilities of FMs in health, facilitating responsible deployment and mitigating associated risks.
comment: Currently under review in IEEE REVIEWS IN BIOMEDICAL ENGINEERING
♻ ☆ Hybrid Approaches for Moral Value Alignment in AI Agents: a Manifesto
Increasing interest in ensuring the safety of next-generation Artificial Intelligence (AI) systems calls for novel approaches to embedding morality into autonomous agents. This goal differs qualitatively from traditional task-specific AI methodologies. In this paper, we provide a systematization of existing approaches to the problem of introducing morality in machines - modelled as a continuum. Our analysis suggests that popular techniques lie at the extremes of this continuum - either being fully hard-coded into top-down, explicit rules, or entirely learned in a bottom-up, implicit fashion with no direct statement of any moral principle (this includes learning from human feedback, as applied to the training and finetuning of large language models, or LLMs). Given the relative strengths and weaknesses of each type of methodology, we argue that more hybrid solutions are needed to create adaptable and robust, yet controllable and interpretable agentic systems. To that end, this paper discusses both the ethical foundations (including deontology, consequentialism and virtue ethics) and implementations of morally aligned AI systems. We present a series of case studies that rely on intrinsic rewards, moral constraints or textual instructions, applied to either pure-Reinforcement Learning or LLM-based agents. By analysing these diverse implementations under one framework, we compare their relative strengths and shortcomings in developing morally aligned AI systems. We then discuss strategies for evaluating the effectiveness of moral learning agents. Finally, we present open research questions and implications for the future of AI safety and ethics which are emerging from this hybrid framework.
♻ ☆ ReFactor GNNs: Revisiting Factorisation-based Models from a Message-Passing Perspective NeurIPS 2022
Factorisation-based Models (FMs), such as DistMult, have enjoyed enduring success for Knowledge Graph Completion (KGC) tasks, often outperforming Graph Neural Networks (GNNs). However, unlike GNNs, FMs struggle to incorporate node features and generalise to unseen nodes in inductive settings. Our work bridges the gap between FMs and GNNs by proposing ReFactor GNNs. This new architecture draws upon both modelling paradigms, which previously were largely thought of as disjoint. Concretely, using a message-passing formalism, we show how FMs can be cast as GNNs by reformulating the gradient descent procedure as message-passing operations, which forms the basis of our ReFactor GNNs. Across a multitude of well-established KGC benchmarks, our ReFactor GNNs achieve comparable transductive performance to FMs, and state-of-the-art inductive performance while using an order of magnitude fewer parameters.
comment: 36th Conference on Neural Information Processing Systems (NeurIPS 2022)
♻ ☆ Local Anti-Concentration Class: Logarithmic Regret for Greedy Linear Contextual Bandit NeurIPS2024
We study the performance guarantees of exploration-free greedy algorithms for the linear contextual bandit problem. We introduce a novel condition, named the \textit{Local Anti-Concentration} (LAC) condition, which enables a greedy bandit algorithm to achieve provable efficiency. We show that the LAC condition is satisfied by a broad class of distributions, including Gaussian, exponential, uniform, Cauchy, and Student's~$t$ distributions, along with other exponential family distributions and their truncated variants. This significantly expands the class of distributions under which greedy algorithms can perform efficiently. Under our proposed LAC condition, we prove that the cumulative expected regret of the greedy algorithm for the linear contextual bandit is bounded by $O(\operatorname{poly} \log T)$. Our results establish the widest range of distributions known to date that allow a sublinear regret bound for greedy algorithms, further achieving a sharp poly-logarithmic regret.
comment: NeurIPS2024
♻ ☆ Higher-Order Topological Directionality and Directed Simplicial Neural Networks
Topological Deep Learning (TDL) has emerged as a paradigm to process and learn from signals defined on higher-order combinatorial topological spaces, such as simplicial or cell complexes. Although many complex systems have an asymmetric relational structure, most TDL models forcibly symmetrize these relationships. In this paper, we first introduce a novel notion of higher-order directionality and we then design Directed Simplicial Neural Networks (Dir-SNNs) based on it. Dir-SNNs are message-passing networks operating on directed simplicial complexes able to leverage directed and possibly asymmetric interactions among the simplices. To our knowledge, this is the first TDL model using a notion of higher-order directionality. We theoretically and empirically prove that Dir-SNNs are more expressive than their directed graph counterpart in distinguishing isomorphic directed graphs. Experiments on a synthetic source localization task demonstrate that Dir-SNNs outperform undirected SNNs when the underlying complex is directed, and perform comparably when the underlying complex is undirected.
comment: 7 pages, 8 figures, 1 table
♻ ☆ Hybrid additive modeling with partial dependence for supervised regression and dynamical systems forecasting
Learning processes by exploiting restricted domain knowledge is an important task across a plethora of scientific areas, with more and more hybrid training methods additively combining data-driven and model-based approaches. Although the obtained models are more accurate than purely data-driven models, the optimization process usually comes with sensitive regularization constraints. Furthermore, while such hybrid methods have been tested in various scientific applications, they have been mostly tested on dynamical systems, with only limited study about the influence of each model component on global performance and parameter identification. In this work, we introduce a new hybrid training approach based on partial dependence, which removes the need for intricate regularization. Moreover, we assess the performance of hybrid modeling against traditional machine learning methods on standard regression problems. We compare, on both synthetic and real regression problems, several approaches for training such hybrid models. We focus on hybrid methods that additively combine a parametric term with a machine learning term and investigate model-agnostic training procedures. Therefore, experiments are carried out with different types of machine learning models, including tree-based models and artificial neural networks. We also extend our partial dependence optimization process for dynamical systems forecasting and compare it to existing schemes.
comment: Extended version of the paper entitled "Knowledge-Guided Additive Modeling for Supervised Regression" (https://link.springer.com/chapter/10.1007/978-3-031-45275-8_5), accepted for publication in the Machine Learning journal. The extension includes new experiments in the static setting, along with a dedicated section on the application of our method to the problem of dynamical systems forecasting
♻ ☆ Provably Efficient Reinforcement Learning with Multinomial Logit Function Approximation NeurIPS 2024
We study a new class of MDPs that employs multinomial logit (MNL) function approximation to ensure valid probability distributions over the state space. Despite its significant benefits, incorporating the non-linear function raises substantial challenges in both statistical and computational efficiency. The best-known result of Hwang and Oh [2023] has achieved an $\widetilde{\mathcal{O}}(\kappa^{-1}dH^2\sqrt{K})$ regret upper bound, where $\kappa$ is a problem-dependent quantity, $d$ is the feature dimension, $H$ is the episode length, and $K$ is the number of episodes. However, we observe that $\kappa^{-1}$ exhibits polynomial dependence on the number of reachable states, which can be as large as the state space size in the worst case and thus undermines the motivation for function approximation. Additionally, their method requires storing all historical data and the time complexity scales linearly with the episode count, which is computationally expensive. In this work, we propose a statistically efficient algorithm that achieves a regret of $\widetilde{\mathcal{O}}(dH^2\sqrt{K} + \kappa^{-1}d^2H^2)$, eliminating the dependence on $\kappa^{-1}$ in the dominant term for the first time. We then address the computational challenges by introducing an enhanced algorithm that achieves the same regret guarantee but with only constant cost. Finally, we establish the first lower bound for this problem, justifying the optimality of our results in $d$ and $K$.
comment: NeurIPS 2024; v3 substantially improves the presentation and further illustrates the role of $\kappa$ in function approximation
♻ ☆ Bayesian Low-Rank LeArning (Bella): A Practical Approach to Bayesian Neural Networks AAAI'2025
Computational complexity of Bayesian learning is impeding its adoption in practical, large-scale tasks. Despite demonstrations of significant merits such as improved robustness and resilience to unseen or out-of-distribution inputs over their non- Bayesian counterparts, their practical use has faded to near insignificance. In this study, we introduce an innovative framework to mitigate the computational burden of Bayesian neural networks (BNNs). Our approach follows the principle of Bayesian techniques based on deep ensembles, but significantly reduces their cost via multiple low-rank perturbations of parameters arising from a pre-trained neural network. Both vanilla version of ensembles as well as more sophisticated schemes such as Bayesian learning with Stein Variational Gradient Descent (SVGD), previously deemed impractical for large models, can be seamlessly implemented within the proposed framework, called Bayesian Low-Rank LeArning (Bella). In a nutshell, i) Bella achieves a dramatic reduction in the number of trainable parameters required to approximate a Bayesian posterior; and ii) it not only maintains, but in some instances, surpasses the performance of conventional Bayesian learning methods and non-Bayesian baselines. Our results with large-scale tasks such as ImageNet, CAMELYON17, DomainNet, VQA with CLIP, LLaVA demonstrate the effectiveness and versatility of Bella in building highly scalable and practical Bayesian deep models for real-world applications.
comment: This paper is accepted in AAAI'2025
♻ ☆ Latent Space Characterization of Autoencoder Variants
Understanding the latent spaces learned by deep learning models is crucial in exploring how they represent and generate complex data. Autoencoders (AEs) have played a key role in the area of representation learning, with numerous regularization techniques and training principles developed not only to enhance their ability to learn compact and robust representations, but also to reveal how different architectures influence the structure and smoothness of the lower-dimensional non-linear manifold. We strive to characterize the structure of the latent spaces learned by different autoencoders including convolutional autoencoders (CAEs), denoising autoencoders (DAEs), and variational autoencoders (VAEs) and how they change with the perturbations in the input. By characterizing the matrix manifolds corresponding to the latent spaces, we provide an explanation for the well-known observation that the latent spaces of CAE and DAE form non-smooth manifolds, while that of VAE forms a smooth manifold. We also map the points of the matrix manifold to a Hilbert space using distance preserving transforms and provide an alternate view in terms of the subspaces generated in the Hilbert space as a function of the distortion in the input. The results show that the latent manifolds of CAE and DAE are stratified with each stratum being a smooth product manifold, while the manifold of VAE is a smooth product manifold of two symmetric positive definite matrices and a symmetric positive semi-definite matrix.
comment: 9 pages, 6 figures, and 1 table
♻ ☆ FSDEM: Feature Selection Dynamic Evaluation Metric
Expressive evaluation metrics are indispensable for informative experiments in all areas, and while several metrics are established in some areas, in others, such as feature selection, only indirect or otherwise limited evaluation metrics are found. In this paper, we propose a novel evaluation metric to address several problems of its predecessors and allow for flexible and reliable evaluation of feature selection algorithms. The proposed metric is a dynamic metric with two properties that can be used to evaluate both the performance and the stability of a feature selection algorithm. We conduct several empirical experiments to illustrate the use of the proposed metric in the successful evaluation of feature selection algorithms. We also provide a comparison and analysis to show the different aspects involved in the evaluation of the feature selection algorithms. The results indicate that the proposed metric is successful in carrying out the evaluation task for feature selection algorithms. This paper is an extended version of a paper published at SISAP 2024.
comment: Short version of this paper is published at 17th International Conference on Similarity Search and Applications, SISAP 2024
♻ ☆ STROOBnet Optimization via GPU-Accelerated Proximal Recurrence Strategies
Spatiotemporal networks' observational capabilities are crucial for accurate data gathering and informed decisions across multiple sectors. This study focuses on the Spatiotemporal Ranged Observer-Observable Bipartite Network (STROOBnet), linking observational nodes (e.g., surveillance cameras) to events within defined geographical regions, enabling efficient monitoring. Using data from Real-Time Crime Camera (RTCC) systems and Calls for Service (CFS) in New Orleans, where RTCC combats rising crime amidst reduced police presence, we address the network's initial observational imbalances. Aiming for uniform observational efficacy, we propose the Proximal Recurrence approach. It outperformed traditional clustering methods like k-means and DBSCAN by offering holistic event frequency and spatial consideration, enhancing observational coverage.
comment: 10 pages, 17 figures, 2023 IEEE International Conference on Big Data (BigData)
♻ ☆ A Consolidated Volatility Prediction with Back Propagation Neural Network and Genetic Algorithm ICML 2024
This paper provides a unique approach with AI algorithms to predict emerging stock markets volatility. Traditionally, stock volatility is derived from historical volatility,Monte Carlo simulation and implied volatility as well. In this paper, the writer designs a consolidated model with back-propagation neural network and genetic algorithm to predict future volatility of emerging stock markets and found that the results are quite accurate with low errors.
comment: 6 pages, 7 figures, 1 table, The paper will be published by IEEE on conference: 2024 3rd International Conference on Image Processing, Computer Vision and Machine Learning (ICICML 2024) (V2)
♻ ☆ SimHawNet: A Modified Hawkes Process for Temporal Network Simulation
Temporal networks allow representing connections between objects while incorporating the temporal dimension. While static network models can capture unchanging topological regularities, they often fail to model the effects associated with the causal generative process of the network that occurs in time. Hence, exploiting the temporal aspect of networks has been the focus of many recent studies. In this context, we propose a new framework for generative models of continuous-time temporal networks. We assume that the activation of the edges in a temporal network is driven by a specified temporal point process. This approach allows to directly model the waiting time between events while incorporating time-varying history-based features as covariates in the predictions. Coupled with a thinning algorithm designed for the simulation of point processes, SimHawNet enables simulation of the evolution of temporal networks in continuous time. Finally, we introduce a comprehensive evaluation framework to assess the performance of such an approach, in which we demonstrate that SimHawNet successfully simulates the evolution of networks with very different generative processes and achieves performance comparable to the state of the art, while being significantly faster.
♻ ☆ Benchmarking quantum machine learning kernel training for classification tasks
Quantum-enhanced machine learning is a rapidly evolving field that aims to leverage the unique properties of quantum mechanics to enhance classical machine learning. However, the practical applicability of these methods remains an open question, particularly beyond the context of specifically-crafted toy problems, and given the current limitations of quantum hardware. This study focuses on quantum kernel methods in the context of classification tasks. In particular, it examines the performance of Quantum Kernel Estimation (QKE) and Quantum Kernel Training (QKT) in connection with two quantum feature mappings, namely ZZFeatureMap and CovariantFeatureMap. Remarkably, these feature maps have been proposed in the literature under the conjecture of possible near-term quantum advantage and have shown promising performance in ad-hoc datasets. In this study, we aim to evaluate their versatility and generalization capabilities in a more general benchmark, encompassing both artificial and established reference datasets. Classical machine learning methods, specifically Support Vector Machines (SVMs) and logistic regression, are also incorporated as baseline comparisons. Experimental results indicate that quantum methods exhibit varying performance across different datasets. Despite outperforming classical methods in ad-hoc datasets, mixed results are obtained for the general case among standard classical benchmarks. Our experiments call into question a general added value of applying QKT optimization, for which the additional computational cost does not necessarily translate into improved classification performance. Instead, it is suggested that a careful choice of the quantum feature map in connection with proper hyperparameterization may prove more effective.
comment: 19 pages, 4 figures; extended experiments and datasets, fixed typos; in consideration for publication in IEEE TQE
♻ ☆ Formation-Controlled Dimensionality Reduction
Dimensionality reduction represents the process of generating a low dimensional representation of high dimensional data. Motivated by the formation control of mobile agents, we propose a nonlinear dynamical system for dimensionality reduction. The system consists of two parts; the control of neighbor points, addressing local structures, and the control of remote points, accounting for global structures.We also include a brief mathematical analysis of the model and its numerical procedure. Numerical experiments are performed on both synthetic and real datasets and comparisons with existing models demonstrate the soundness and effectiveness of the proposed model.
♻ ☆ Sparsity-Aware Distributed Learning for Gaussian Processes with Linear Multiple Kernel
Gaussian processes (GPs) stand as crucial tools in machine learning and signal processing, with their effectiveness hinging on kernel design and hyper-parameter optimization. This paper presents a novel GP linear multiple kernel (LMK) and a generic sparsity-aware distributed learning framework to optimize the hyper-parameters. The newly proposed grid spectral mixture product (GSMP) kernel is tailored for multi-dimensional data, effectively reducing the number of hyper-parameters while maintaining good approximation capability. We further demonstrate that the associated hyper-parameter optimization of this kernel yields sparse solutions. To exploit the inherent sparsity of the solutions, we introduce the Sparse LInear Multiple Kernel Learning (SLIM-KL) framework. The framework incorporates a quantized alternating direction method of multipliers (ADMM) scheme for collaborative learning among multiple agents, where the local optimization problem is solved using a distributed successive convex approximation (DSCA) algorithm. SLIM-KL effectively manages large-scale hyper-parameter optimization for the proposed kernel, simultaneously ensuring data privacy and minimizing communication costs. Theoretical analysis establishes convergence guarantees for the learning framework, while experiments on diverse datasets demonstrate the superior prediction performance and efficiency of our proposed methods.
♻ ☆ AALF: Almost Always Linear Forecasting
Recent works for time-series forecasting more and more leverage the high predictive power of Deep Learning models. With this increase in model complexity, however, comes a lack in understanding of the underlying model decision process, which is problematic for high-stakes application scenarios. At the same time, simple, interpretable forecasting methods such as ARIMA still perform very well, sometimes on-par, with Deep Learning approaches. We argue that simple models are good enough most of the time, and that forecasting performance could be improved by choosing a Deep Learning method only for few, important predictions, increasing the overall interpretability of the forecasting process. In this context, we propose a novel online model selection framework which learns to identify these predictions. An extensive empirical study on various real-world datasets shows that our selection methodology performs comparable to state-of-the-art online model selections methods in most cases while being significantly more interpretable. We find that almost always choosing a simple autoregressive linear model for forecasting results in competitive performance, suggesting that the need for opaque black-box models in time-series forecasting might be smaller than recent works would suggest.
♻ ☆ Evaluating alignment between humans and neural network representations in image-based learning tasks
Humans represent scenes and objects in rich feature spaces, carrying information that allows us to generalise about category memberships and abstract functions with few examples. What determines whether a neural network model generalises like a human? We tested how well the representations of $86$ pretrained neural network models mapped to human learning trajectories across two tasks where humans had to learn continuous relationships and categories of natural images. In these tasks, both human participants and neural networks successfully identified the relevant stimulus features within a few trials, demonstrating effective generalisation. We found that while training dataset size was a core determinant of alignment with human choices, contrastive training with multi-modal data (text and imagery) was a common feature of currently publicly available models that predicted human generalisation. Intrinsic dimensionality of representations had different effects on alignment for different model types. Lastly, we tested three sets of human-aligned representations and found no consistent improvements in predictive accuracy compared to the baselines. In conclusion, pretrained neural networks can serve to extract representations for cognitive models, as they appear to capture some fundamental aspects of cognition that are transferable across tasks. Both our paradigms and modelling approach offer a novel way to quantify alignment between neural networks and humans and extend cognitive science into more naturalistic domains.
♻ ☆ Instruction-Guided Fusion of Multi-Layer Visual Features in Large Vision-Language Models
Large Vision-Language Models (LVLMs) have achieved significant success in multimodal tasks by combining pre-trained vision encoders and large language models. However, current LVLMs mainly rely on features from the final layers of the vision encoder, neglecting complementary information in shallower layers. While recent methods have explored multi-layer features, they are often task-agnostic. We investigate the contributions of visual features from different encoder layers across 18 benchmarks and 6 task categories. Our results show that multi-layer features provide complementary strengths with varying task dependencies, and uniform fusion performs suboptimally. Based on these findings, we propose an instruction-guided vision aggregator that dynamically integrates multi-layer features based on textual instructions, without increasing the number of visual tokens. Extensive evaluations show superior performance, and analysis reveals the dominance of mid-to-high-level features in semantic tasks and the critical role of low-level features in fine-grained perception. This work provides valuable insights into the adaptive use of hierarchical visual features in LVLMs, advancing more flexible multimodal systems.
♻ ☆ Wasserstein Gradient Flows for Moreau Envelopes of f-Divergences in Reproducing Kernel Hilbert Spaces
Commonly used $f$-divergences of measures, e.g., the Kullback-Leibler divergence, are subject to limitations regarding the support of the involved measures. A remedy is regularizing the $f$-divergence by a squared maximum mean discrepancy (MMD) associated with a characteristic kernel $K$. We use the kernel mean embedding to show that this regularization can be rewritten as the Moreau envelope of some function on the associated reproducing kernel Hilbert space. Then, we exploit well-known results on Moreau envelopes in Hilbert spaces to analyze the MMD-regularized $f$-divergences, particularly their gradients. Subsequently, we use our findings to analyze Wasserstein gradient flows of MMD-regularized $f$-divergences. We provide proof-of-the-concept numerical examples for flows starting from empirical measures. Here, we cover $f$-divergences with infinite and finite recession constants. Lastly, we extend our results to the tight variational formulation of $f$-divergences and numerically compare the resulting flows.
comment: 56 pages, 14 figures, 3 tables. Comments welcome! NEW: Incorporated Reviewers' suggestions, added FISTA and tight formulation
♻ ☆ Learning Constraint Network from Demonstrations via Positive-Unlabeled Learning with Memory Replay
Planning for a wide range of real-world tasks necessitates to know and write all constraints. However, instances exist where these constraints are either unknown or challenging to specify accurately. A possible solution is to infer the unknown constraints from expert demonstration. The majority of prior works limit themselves to learning simple linear constraints, or require strong knowledge of the true constraint parameterization or environmental model. To mitigate these problems, this paper presents a positive-unlabeled (PU) learning approach to infer a continuous, arbitrary and possibly nonlinear, constraint from demonstration. From a PU learning view, We treat all data in demonstrations as positive (feasible) data, and learn a (sub)-optimal policy to generate high-reward-winning but potentially infeasible trajectories, which serve as unlabeled data containing both feasible and infeasible states. Under an assumption on data distribution, a feasible-infeasible classifier (i.e., constraint model) is learned from the two datasets through a postprocessing PU learning technique. The entire method employs an iterative framework alternating between updating the policy, which generates and selects higher-reward policies, and updating the constraint model. Additionally, a memory buffer is introduced to record and reuse samples from previous iterations to prevent forgetting. The effectiveness of the proposed method is validated in two Mujoco environments, successfully inferring continuous nonlinear constraints and outperforming a baseline method in terms of constraint accuracy and policy safety.
♻ ☆ Focus On This, Not That! Steering LLMs With Adaptive Feature Specification
Despite the success of Instruction Tuning (IT) in training large language models (LLMs) to perform arbitrary user-specified tasks, these models often still leverage spurious or biased features learned from their training data, leading to undesired behaviours when deploying them in new contexts. In this work, we introduce Focus Instruction Tuning (FIT), which trains LLMs to condition their responses by focusing on specific features whilst ignoring others, leading to different behaviours based on what features are specified. Across several experimental settings, we show that focus-tuned models can be adaptively steered by focusing on different features at inference-time: for instance, robustness can be improved by focusing on task-causal features and ignoring spurious features, and social bias can be mitigated by ignoring demographic categories. Furthermore, FIT can steer behaviour in new contexts, generalising under distribution shift and to new unseen features at inference time, and thereby facilitating more robust, fair, and controllable LLM applications in real-world environments.
comment: 28pages, 14 figures
♻ ☆ Diffusion Models in Vision: A Survey
Denoising diffusion models represent a recent emerging topic in computer vision, demonstrating remarkable results in the area of generative modeling. A diffusion model is a deep generative model that is based on two stages, a forward diffusion stage and a reverse diffusion stage. In the forward diffusion stage, the input data is gradually perturbed over several steps by adding Gaussian noise. In the reverse stage, a model is tasked at recovering the original input data by learning to gradually reverse the diffusion process, step by step. Diffusion models are widely appreciated for the quality and diversity of the generated samples, despite their known computational burdens, i.e. low speeds due to the high number of steps involved during sampling. In this survey, we provide a comprehensive review of articles on denoising diffusion models applied in vision, comprising both theoretical and practical contributions in the field. First, we identify and present three generic diffusion modeling frameworks, which are based on denoising diffusion probabilistic models, noise conditioned score networks, and stochastic differential equations. We further discuss the relations between diffusion models and other deep generative models, including variational auto-encoders, generative adversarial networks, energy-based models, autoregressive models and normalizing flows. Then, we introduce a multi-perspective categorization of diffusion models applied in computer vision. Finally, we illustrate the current limitations of diffusion models and envision some interesting directions for future research.
comment: Accepted in IEEE Transactions on Pattern Analysis and Machine Intelligence. 25 pages, 3 figures
♻ ☆ WindsorML: High-Fidelity Computational Fluid Dynamics Dataset For Automotive Aerodynamics
This paper presents a new open-source high-fidelity dataset for Machine Learning (ML) containing 355 geometric variants of the Windsor body, to help the development and testing of ML surrogate models for external automotive aerodynamics. Each Computational Fluid Dynamics (CFD) simulation was run with a GPU-native high-fidelity Wall-Modeled Large-Eddy Simulations (WMLES) using a Cartesian immersed-boundary method using more than 280M cells to ensure the greatest possible accuracy. The dataset contains geometry variants that exhibits a wide range of flow characteristics that are representative of those observed on road-cars. The dataset itself contains the 3D time-averaged volume & boundary data as well as the geometry and force & moment coefficients. This paper discusses the validation of the underlying CFD methods as well as contents and structure of the dataset. To the authors knowledge, this represents the first, large-scale high-fidelity CFD dataset for the Windsor body with a permissive open-source license (CC-BY-SA).
♻ ☆ Contrastive Policy Gradient: Aligning LLMs on sequence-level scores in a supervised-friendly fashion EMNLP 2024
Reinforcement Learning (RL) has been used to finetune Large Language Models (LLMs) using a reward model trained from preference data, to better align with human judgment. The recently introduced direct alignment methods, which are often simpler, more stable, and computationally lighter, can more directly achieve this. However, these approaches cannot optimize arbitrary rewards, and the preference-based ones are not the only rewards of interest for LLMs (eg., unit tests for code generation or textual entailment for summarization, among others). RL-finetuning is usually done with a variation of policy gradient, which calls for on-policy or near-on-policy samples, requiring costly generations. We introduce Contrastive Policy Gradient, or CoPG, a simple and mathematically principled new RL algorithm that can estimate the optimal policy even from off-policy data. It can be seen as an off-policy policy gradient approach that does not rely on important sampling techniques and highlights the importance of using (the right) state baseline. We show this approach to generalize the direct alignment method IPO (identity preference optimization) and classic policy gradient. We experiment with the proposed CoPG on a toy bandit problem to illustrate its properties, as well as for finetuning LLMs on a summarization task, using a learned reward function considered as ground truth for the purpose of the experiments.
comment: EMNLP 2024
♻ ☆ Dataset-Free Weight-Initialization on Restricted Boltzmann Machine
In feed-forward neural networks, dataset-free weight-initialization methods such as LeCun, Xavier (or Glorot), and He initializations have been developed. These methods randomly determine the initial values of weight parameters based on specific distributions (e.g., Gaussian or uniform distributions) without using training datasets. To the best of the authors' knowledge, such a dataset-free weight-initialization method is yet to be developed for restricted Boltzmann machines (RBMs), which are probabilistic neural networks consisting of two layers. In this study, we derive a dataset-free weight-initialization method for Bernoulli--Bernoulli RBMs based on statistical mechanical analysis. In the proposed weight-initialization method, the weight parameters are drawn from a Gaussian distribution with zero mean. The standard deviation of the Gaussian distribution is optimized based on our hypothesis that a standard deviation providing a larger layer correlation (LC) between the two layers improves the learning efficiency. The expression of the LC is derived based on a statistical mechanical analysis. The optimal value of the standard deviation corresponds to the maximum point of the LC. The proposed weight-initialization method is identical to Xavier initialization in a specific case (i.e., when the sizes of the two layers are the same, the random variables of the layers are $\{-1,1\}$-binary, and all bias parameters are zero). The validity of the proposed weight-initialization method is demonstrated in numerical experiments using a toy and real-world datasets.
♻ ☆ Machine Learning Nonadiabatic Dynamics: Eliminating Phase Freedom of Nonadiabatic Couplings with the State-Intraction State-Averaged Spin-Restricted Ensemble-Referenced Kohn-Sham Approach
Excited-state molecular dynamics (ESMD) simulations near conical intersections (CIs) pose significant challenges when using machine learning potentials (MLPs). Although MLPs have gained recognition for their integration into mixed quantum-classical (MQC) methods, such as trajectory surface hopping (TSH), and their capacity to model correlated electron-nuclear dynamics efficiently, difficulties persist in managing nonadiabatic dynamics. Specifically, singularities at CIs and double-valued coupling elements result in discontinuities that disrupt the smoothness of predictive functions. Partial solutions have been provided by learning diabatic Hamiltonians with phaseless loss functions to these challenges. However, a definitive method for addressing the discontinuities caused by CIs and double-valued coupling elements has yet to be developed. Here, we introduce the phaseless coupling term, $\Delta^2$, derived from the square of the off-diagonal elements of the diabatic Hamiltonian in the state-interaction state-averaged spin-restricted ensemble-referenced Kohn-Sham (SI-SA-REKS, briefly SSR)(2,2) formalism. This approach improves the stability and accuracy of the MLP model by addressing the issues arising from CI singularities and double-valued coupling functions. We apply this method to the penta-2,4-dieniminium cation (PSB3), demonstrating its effectiveness in improving MLP training for ML-based nonadiabatic dynamics. Our results show that the $\Delta^2$ based ML-ESMD method can reproduce ab initio ESMD simulations, underscoring its potential and efficiency for broader applications, particularly in large-scale and long-timescale ESMD simulations.
♻ ☆ Positive-Unlabeled Constraint Learning for Inferring Nonlinear Continuous Constraints Functions from Expert Demonstrations
Planning for diverse real-world robotic tasks necessitates to know and write all constraints. However, instances exist where these constraints are either unknown or challenging to specify accurately. A possible solution is to infer the unknown constraints from expert demonstration. This paper presents a novel two-step Positive-Unlabeled Constraint Learning (PUCL) algorithm to infer a continuous constraint function from demonstrations, without requiring prior knowledge of the true constraint parameterization or environmental model as existing works. We treat all data in demonstrations as positive (feasible) data, and learn a control policy to generate potentially infeasible trajectories, which serve as unlabeled data. The proposed two-step learning framework first identifies reliable infeasible data using a distance metric, and secondly learns a binary feasibility classifier (i.e., constraint function) from the feasible demonstrations and reliable infeasible data. The proposed method is flexible to learn complex-shaped constraint boundary and will not mistakenly classify demonstrations as infeasible as previous methods. The effectiveness of the proposed method is verified in four constrained environments, using a networked policy or a dynamical system policy. It successfully infers the continuous nonlinear constraints and outperforms other baseline methods in terms of constraint accuracy and policy safety. This work has been published in IEEE Robotics and Automation Letters (RA-L). Please refer to the final version at https://doi.org/10.1109/LRA.2024.3522756
♻ ☆ ERGNN: Spectral Graph Neural Network With Explicitly-Optimized Rational Graph Filters ICASSP 2025
Approximation-based spectral graph neural networks, which construct graph filters with function approximation, have shown substantial performance in graph learning tasks. Despite their great success, existing works primarily employ polynomial approximation to construct the filters, whereas another superior option, namely ration approximation, remains underexplored. Although a handful of prior works have attempted to deploy the rational approximation, their implementations often involve intensive computational demands or still resort to polynomial approximations, hindering full potential of the rational graph filters. To address the issues, this paper introduces ERGNN, a novel spectral GNN with explicitly-optimized rational filter. ERGNN adopts a unique two-step framework that sequentially applies the numerator filter and the denominator filter to the input signals, thus streamlining the model paradigm while enabling explicit optimization of both numerator and denominator of the rational filter. Extensive experiments validate the superiority of ERGNN over state-of-the-art methods, establishing it as a practical solution for deploying rational-based GNNs.
comment: Accepted in 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025
♻ ☆ An Adaptive Collocation Point Strategy For Physics Informed Neural Networks via the QR Discrete Empirical Interpolation Method ICML 2025
Physics-informed neural networks (PINNs) have gained significant attention for solving forward and inverse problems related to partial differential equations (PDEs). While advancements in loss functions and network architectures have improved PINN accuracy, the impact of collocation point sampling on their performance remains underexplored. Fixed sampling methods, such as uniform random sampling and equispaced grids, can fail to capture critical regions with high solution gradients, limiting their effectiveness for complex PDEs. Adaptive methods, inspired by adaptive mesh refinement from traditional numerical methods, address this by dynamically updating collocation points during training but may overlook residual dynamics between updates, potentially losing valuable information. To overcome this limitation, we propose an adaptive collocation point selection strategy utilizing the QR Discrete Empirical Interpolation Method (QR-DEIM), a reduced-order modeling technique for efficiently approximating nonlinear functions. Our results on benchmark PDEs, including the wave, Allen-Cahn, and Burgers' equations, demonstrate that our QR-DEIM-based approach improves PINN accuracy compared to existing methods, offering a promising direction for adaptive collocation point strategies.
comment: Submitted to ICML 2025. Under review
♻ ☆ Deterministic Uncertainty Propagation for Improved Model-Based Offline Reinforcement Learning
Current approaches to model-based offline reinforcement learning often incorporate uncertainty-based reward penalization to address the distributional shift problem. These approaches, commonly known as pessimistic value iteration, use Monte Carlo sampling to estimate the Bellman target to perform temporal difference-based policy evaluation. We find out that the randomness caused by this sampling step significantly delays convergence. We present a theoretical result demonstrating the strong dependency of suboptimality on the number of Monte Carlo samples taken per Bellman target calculation. Our main contribution is a deterministic approximation to the Bellman target that uses progressive moment matching, a method developed originally for deterministic variational inference. The resulting algorithm, which we call Moment Matching Offline Model-Based Policy Optimization (MOMBO), propagates the uncertainty of the next state through a nonlinear Q-network in a deterministic fashion by approximating the distributions of hidden layer activations by a normal distribution. We show that it is possible to provide tighter guarantees for the suboptimality of MOMBO than the existing Monte Carlo sampling approaches. We also observe MOMBO to converge faster than these approaches in a large set of benchmark tasks.
♻ ☆ Disentangled Interleaving Variational Encoding
Conflicting objectives present a considerable challenge in interleaving multi-task learning, necessitating the need for meticulous design and balance to ensure effective learning of a representative latent data space across all tasks without mutual negative impact. Drawing inspiration from the concept of marginal and conditional probability distributions in probability theory, we design a principled and well-founded approach to disentangle the original input into marginal and conditional probability distributions in the latent space of a variational autoencoder. Our proposed model, Deep Disentangled Interleaving Variational Encoding (DeepDIVE) learns disentangled features from the original input to form clusters in the embedding space and unifies these features via the cross-attention mechanism in the fusion stage. We theoretically prove that combining the objectives for reconstruction and forecasting fully captures the lower bound and mathematically derive a loss function for disentanglement using Na\"ive Bayes. Under the assumption that the prior is a mixture of log-concave distributions, we also establish that the Kullback-Leibler divergence between the prior and the posterior is upper bounded by a function minimized by the minimizer of the cross entropy loss, informing our adoption of radial basis functions (RBF) and cross entropy with interleaving training for DeepDIVE to provide a justified basis for convergence. Experiments on two public datasets show that DeepDIVE disentangles the original input and yields forecast accuracies better than the original VAE and comparable to existing state-of-the-art baselines.
♻ ☆ PeFLL: Personalized Federated Learning by Learning to Learn
We present PeFLL, a new personalized federated learning algorithm that improves over the state-of-the-art in three aspects: 1) it produces more accurate models, especially in the low-data regime, and not only for clients present during its training phase, but also for any that may emerge in the future; 2) it reduces the amount of on-client computation and client-server communication by providing future clients with ready-to-use personalized models that require no additional finetuning or optimization; 3) it comes with theoretical guarantees that establish generalization from the observed clients to future ones. At the core of PeFLL lies a learning-to-learn approach that jointly trains an embedding network and a hypernetwork. The embedding network is used to represent clients in a latent descriptor space in a way that reflects their similarity to each other. The hypernetwork takes as input such descriptors and outputs the parameters of fully personalized client models. In combination, both networks constitute a learning algorithm that achieves state-of-the-art performance in several personalized federated learning benchmarks.
♻ ☆ LLM360 K2: Building a 65B 360-Open-Source Large Language Model from Scratch
We detail the training of the LLM360 K2-65B model, scaling up our 360-degree OPEN SOURCE approach to the largest and most powerful models under project LLM360. While open-source LLMs continue to advance, the answer to "How are the largest LLMs trained?" remains unclear within the community. The implementation details for such high-capacity models are often protected due to business considerations associated with their high cost. This lack of transparency prevents LLM researchers from leveraging valuable insights from prior experience, e.g., "What are the best practices for addressing loss spikes?" The LLM360 K2 project addresses this gap by providing full transparency and access to resources accumulated during the training of LLMs at the largest scale. This report highlights key elements of the K2 project, including our first model, K2 DIAMOND, a 65 billion-parameter LLM that surpasses LLaMA-65B and rivals LLaMA2-70B, while requiring fewer FLOPs and tokens. We detail the implementation steps and present a longitudinal analysis of K2 DIAMOND's capabilities throughout its training process. We also outline ongoing projects such as TXT360, setting the stage for future models in the series. By offering previously unavailable resources, the K2 project also resonates with the 360-degree OPEN SOURCE principles of transparency, reproducibility, and accessibility, which we believe are vital in the era of resource-intensive AI research.
♻ ☆ Enhancing User Interest based on Stream Clustering and Memory Networks in Large-Scale Recommender Systems
Recommender Systems (RSs) provide personalized recommendation service based on user interest, which are widely used in various platforms. However, there are lots of users with sparse interest due to lacking consumption behaviors, which leads to poor recommendation results for them. This problem is widespread in large-scale RSs and is particularly difficult to address. To solve this problem, we propose a novel solution named User Interest Enhancement (UIE) which enhances user interest including user profile and user history behavior sequences using the enhancement vectors and personalized enhancement vector generated based on stream clustering and memory networks from different perspectives. UIE not only remarkably improves model performance on the users with sparse interest but also significantly enhance model performance on other users. UIE is an end-to-end solution which is easy to be implemented based on ranking model. Moreover, we expand our solution and apply similar methods to long-tail items, which also achieves excellent improvement. Furthermore, we conduct extensive offline and online experiments in a large-scale industrial RS. The results demonstrate that our model outperforms other models remarkably, especially for the users with sparse interest. Until now, UIE has been fully deployed in multiple large-scale RSs and achieved remarkable improvements.
♻ ☆ Simplified and Generalized Masked Diffusion for Discrete Data NeurIPS 2024
Masked (or absorbing) diffusion is actively explored as an alternative to autoregressive models for generative modeling of discrete data. However, existing work in this area has been hindered by unnecessarily complex model formulations and unclear relationships between different perspectives, leading to suboptimal parameterization, training objectives, and ad hoc adjustments to counteract these issues. In this work, we aim to provide a simple and general framework that unlocks the full potential of masked diffusion models. We show that the continuous-time variational objective of masked diffusion models is a simple weighted integral of cross-entropy losses. Our framework also enables training generalized masked diffusion models with state-dependent masking schedules. When evaluated by perplexity, our models trained on OpenWebText surpass prior diffusion language models at GPT-2 scale and demonstrate superior performance on 4 out of 5 zero-shot language modeling tasks. Furthermore, our models vastly outperform previous discrete diffusion models on pixel-level image modeling, achieving 2.75 (CIFAR-10) and 3.40 (ImageNet 64x64) bits per dimension that are better than autoregressive models of similar sizes. Our code is available at https://github.com/google-deepmind/md4.
comment: NeurIPS 2024. Code is available at: https://github.com/google-deepmind/md4
♻ ☆ Balancing Act: Prioritization Strategies for LLM-Designed Restless Bandit Rewards
LLMs are increasingly used to design reward functions based on human preferences in Reinforcement Learning (RL). We focus on LLM-designed rewards for Restless Multi-Armed Bandits, a framework for allocating limited resources among agents. In applications such as public health, this approach empowers grassroots health workers to tailor automated allocation decisions to community needs. In the presence of multiple agents, altering the reward function based on human preferences can impact subpopulations very differently, leading to complex tradeoffs and a multi-objective resource allocation problem. We are the first to present a principled method termed Social Choice Language Model for dealing with these tradeoffs for LLM-designed rewards for multiagent planners in general and restless bandits in particular. The novel part of our model is a transparent and configurable selection component, called an adjudicator, external to the LLM that controls complex tradeoffs via a user-selected social welfare function. Our experiments demonstrate that our model reliably selects more effective, aligned, and balanced reward functions compared to purely LLM-based approaches.
♻ ☆ Learning to Assist Humans without Inferring Rewards NeurIPS
Assistive agents should make humans' lives easier. Classically, such assistance is studied through the lens of inverse reinforcement learning, where an assistive agent (e.g., a chatbot, a robot) infers a human's intention and then selects actions to help the human reach that goal. This approach requires inferring intentions, which can be difficult in high-dimensional settings. We build upon prior work that studies assistance through the lens of empowerment: an assistive agent aims to maximize the influence of the human's actions such that they exert a greater control over the environmental outcomes and can solve tasks in fewer steps. We lift the major limitation of prior work in this area--scalability to high-dimensional settings--with contrastive successor representations. We formally prove that these representations estimate a similar notion of empowerment to that studied by prior work and provide a ready-made mechanism for optimizing it. Empirically, our proposed method outperforms prior methods on synthetic benchmarks, and scales to Overcooked, a cooperative game setting. Theoretically, our work connects ideas from information theory, neuroscience, and reinforcement learning, and charts a path for representations to play a critical role in solving assistive problems.
comment: Conference on Neural Information Processing Systems (NeurIPS), 2024
♻ ☆ Discriminative Representation learning via Attention-Enhanced Contrastive Learning for Short Text Clustering
Contrastive learning has gained significant attention in short text clustering, yet it has an inherent drawback of mistakenly identifying samples from the same category as negatives and then separating them in the feature space (false negative separation), which hinders the generation of superior representations. To generate more discriminative representations for efficient clustering, we propose a novel short text clustering method, called Discriminative Representation learning via \textbf{A}ttention-\textbf{E}nhanced \textbf{C}ontrastive \textbf{L}earning for Short Text Clustering (\textbf{AECL}). The \textbf{AECL} consists of two modules which are the pseudo-label generation module and the contrastive learning module. Both modules build a sample-level attention mechanism to capture similarity relationships between samples and aggregate cross-sample features to generate consistent representations. Then, the former module uses the more discriminative consistent representation to produce reliable supervision information for assist clustering, while the latter module explores similarity relationships and consistent representations optimize the construction of positive samples to perform similarity-guided contrastive learning, effectively addressing the false negative separation issue. Experimental results demonstrate that the proposed \textbf{AECL} outperforms state-of-the-art methods. If the paper is accepted, we will open-source the code.
♻ ☆ PsyDI: Towards a Personalized and Progressively In-depth Chatbot for Psychological Measurements
In the field of psychology, traditional assessment methods, such as standardized scales, are frequently critiqued for their static nature, lack of personalization, and reduced participant engagement, while comprehensive counseling evaluations are often inaccessible. The complexity of quantifying psychological traits further limits these methods. Despite advances with large language models (LLMs), many still depend on single-round Question-and-Answer interactions. To bridge this gap, we introduce PsyDI, a personalized and progressively in-depth chatbot designed for psychological measurements, exemplified by its application in the Myers-Briggs Type Indicator (MBTI) framework. PsyDI leverages user-related multi-modal information and engages in customized, multi-turn interactions to provide personalized, easily accessible measurements, while ensuring precise MBTI type determination. To address the challenge of unquantifiable psychological traits, we introduce a novel training paradigm that involves learning the ranking of proxy variables associated with these traits, culminating in a robust score model for MBTI measurements. The score model enables PsyDI to conduct comprehensive and precise measurements through multi-turn interactions within a unified estimation context. Through various experiments, we validate the efficacy of both the score model and the PsyDI pipeline, demonstrating its potential to serve as a general framework for psychological measurements. Furthermore, the online deployment of PsyDI has garnered substantial user engagement, with over 3,000 visits, resulting in the collection of numerous multi-turn dialogues annotated with MBTI types, which facilitates further research. The source code for the training and web service components is publicly available as a part of OpenDILab at: https://github.com/opendilab/PsyDI
comment: 29 pages, 15 figures
♻ ☆ Mitigating Overfitting in Graph Neural Networks via Feature and Hyperplane Perturbation
Graph neural networks (GNNs) are commonly used in semi-supervised settings. Previous research has primarily focused on finding appropriate graph filters (e.g. aggregation methods) to perform well on both homophilic and heterophilic graphs. While these methods are effective, they can still suffer from the sparsity of node features, where the initial data contain few non-zero elements. This can lead to overfitting in certain dimensions in the first projection matrix, as training samples may not cover the entire range of graph filters (hyperplanes). To address this, we propose a novel data augmentation strategy. Specifically, by flipping both the initial features and hyperplane, we create additional space for training, which leads to more precise updates of the learnable parameters and improved robustness for unseen features during inference. To the best of our knowledge, this is the first attempt to mitigate the overfitting caused by the initial features. Extensive experiments on real-world datasets show that our proposed technique increases node classification accuracy by up to 46.5% relatively.
♻ ☆ Enhanced SPS Velocity-adaptive Scheme: Access Fairness in 5G NR V2I Networks SP
Vehicle-to-Infrastructure (V2I) technology enables information exchange between vehicles and road infrastructure. Specifically, when a vehicle approaches a roadside unit (RSU), it can exchange information with the RSU to obtain accurate data that assists in driving. With the release of the 3rd Generation Partnership Project (3GPP) Release 16, which includes the 5G New Radio (NR) Vehicle-to-Everything (V2X) standards, vehicles typically adopt mode-2 communication using sensing-based semi-persistent scheduling (SPS) for resource allocation. In this approach, vehicles identify candidate resources within a selection window and exclude ineligible resources based on information from a sensing window. However, vehicles often drive at different speeds, resulting in varying amounts of data transmission with RSUs as they pass by, which leads to unfair access. Therefore, it is essential to design an access scheme that accounts for different vehicle speeds to achieve fair access across the network. This paper formulates an optimization problem for vehicular networks and proposes a multi-objective optimization scheme to address it by adjusting the selection window in the SPS mechanism of 5G NR V2I mode-2. Simulation results demonstrate the effectiveness of the proposed scheme
comment: This paper has been submitted to IEEE Journal. The source code has been released at: https://github.com/qiongwu86/Enhanced-SPS-Velocity-adaptiveScheme-Access-Fariness-in-5G-NR-V2I-Networks
♻ ☆ Efficient PAC Learning of Halfspaces with Constant Malicious Noise Rate ALT 2025
Understanding noise tolerance of machine learning algorithms is a central quest in learning theory. In this work, we study the problem of computationally efficient PAC learning of halfspaces in the presence of malicious noise, where an adversary can corrupt both instances and labels of training samples. The best-known noise tolerance either depends on a target error rate under distributional assumptions or on a margin parameter under large-margin conditions. In this work, we show that when both types of conditions are satisfied, it is possible to achieve constant noise tolerance by minimizing a reweighted hinge loss. Our key ingredients include: 1) an efficient algorithm that finds weights to control the gradient deterioration from corrupted samples, and 2) a new analysis on the robustness of the hinge loss equipped with such weights.
comment: ALT 2025
♻ ☆ VLG-CBM: Training Concept Bottleneck Models with Vision-Language Guidance NeurIPS 2024
Concept Bottleneck Models (CBMs) provide interpretable prediction by introducing an intermediate Concept Bottleneck Layer (CBL), which encodes human-understandable concepts to explain models' decision. Recent works proposed to utilize Large Language Models and pre-trained Vision-Language Models to automate the training of CBMs, making it more scalable and automated. However, existing approaches still fall short in two aspects: First, the concepts predicted by CBL often mismatch the input image, raising doubts about the faithfulness of interpretation. Second, it has been shown that concept values encode unintended information: even a set of random concepts could achieve comparable test accuracy to state-of-the-art CBMs. To address these critical limitations, in this work, we propose a novel framework called Vision-Language-Guided Concept Bottleneck Model (VLG-CBM) to enable faithful interpretability with the benefits of boosted performance. Our method leverages off-the-shelf open-domain grounded object detectors to provide visually grounded concept annotation, which largely enhances the faithfulness of concept prediction while further improving the model performance. In addition, we propose a new metric called Number of Effective Concepts (NEC) to control the information leakage and provide better interpretability. Extensive evaluations across five standard benchmarks show that our method, VLG-CBM, outperforms existing methods by at least 4.27% and up to 51.09% on Accuracy at NEC=5 (denoted as ANEC-5), and by at least 0.45% and up to 29.78% on average accuracy (denoted as ANEC-avg), while preserving both faithfulness and interpretability of the learned concepts as demonstrated in extensive experiments.
comment: Appeared at NeurIPS 2024
♻ ☆ The Power of Types: Exploring the Impact of Type Checking on Neural Bug Detection in Dynamically Typed Languages ICSE'25
Motivation: Automated bug detection in dynamically typed languages such as Python is essential for maintaining code quality. The lack of mandatory type annotations in such languages can lead to errors that are challenging to identify early with traditional static analysis tools. Recent progress in deep neural networks has led to increased use of neural bug detectors. In statically typed languages, a type checker is integrated into the compiler and thus taken into consideration when the neural bug detector is designed for these languages. Problem: However, prior studies overlook this aspect during the training and testing of neural bug detectors for dynamically typed languages. When an optional type checker is used, assessing existing neural bug detectors on bugs easily detectable by type checkers may impact their performance estimation. Moreover, including these bugs in the training set of neural bug detectors can shift their detection focus toward the wrong type of bugs. Contribution: We explore the impact of type checking on various neural bug detectors for variable misuse bugs, a common type targeted by neural bug detectors. Existing synthetic and real-world datasets are type-checked to evaluate the prevalence of type-related bugs. Then, we investigate how type-related bugs influence the training and testing of the neural bug detectors. Findings: Our findings indicate that existing bug detection datasets contain a significant proportion of type-related bugs. Building on this insight, we discover integrating the neural bug detector with a type checker can be beneficial, especially when the code is annotated with types. Further investigation reveals neural bug detectors perform better on type-related bugs than other bugs. Moreover, removing type-related bugs from the training data helps improve neural bug detectors' ability to identify bugs beyond the scope of type checkers.
comment: Accepted by ICSE'25 Research Track
♻ ☆ Cost-aware Bayesian Optimization via the Pandora's Box Gittins Index
Bayesian optimization is a technique for efficiently optimizing unknown functions in a black-box manner. To handle practical settings where gathering data requires use of finite resources, it is desirable to explicitly incorporate function evaluation costs into Bayesian optimization policies. To understand how to do so, we develop a previously-unexplored connection between cost-aware Bayesian optimization and the Pandora's Box problem, a decision problem from economics. The Pandora's Box problem admits a Bayesian-optimal solution based on an expression called the Gittins index, which can be reinterpreted as an acquisition function. We study the use of this acquisition function for cost-aware Bayesian optimization, and demonstrate empirically that it performs well, particularly in medium-high dimensions. We further show that this performance carries over to classical Bayesian optimization without explicit evaluation costs. Our work constitutes a first step towards integrating techniques from Gittins index theory into Bayesian optimization.
♻ ☆ Graph Analysis Using a GPU-based Parallel Algorithm: Quantum Clustering
The article introduces a new method for applying Quantum Clustering to graph structures. Quantum Clustering (QC) is a novel density-based unsupervised learning method that determines cluster centers by constructing a potential function. In this method, we use the Graph Gradient Descent algorithm to find the centers of clusters. GPU parallelization is utilized for computing potential values. We also conducted experiments on five widely used datasets and evaluated using four indicators. The results show superior performance of the method. Finally, we discuss the influence of $\sigma$ on the experimental results.
♻ ☆ The surprising efficiency of temporal difference learning for rare event prediction NeurIPS 2024
We quantify the efficiency of temporal difference (TD) learning over the direct, or Monte Carlo (MC), estimator for policy evaluation in reinforcement learning, with an emphasis on estimation of quantities related to rare events. Policy evaluation is complicated in the rare event setting by the long timescale of the event and by the need for \emph{relative accuracy} in estimates of very small values. Specifically, we focus on least-squares TD (LSTD) prediction for finite state Markov chains, and show that LSTD can achieve relative accuracy far more efficiently than MC. We prove a central limit theorem for the LSTD estimator and upper bound the \emph{relative asymptotic variance} by simple quantities characterizing the connectivity of states relative to the transition probabilities between them. Using this bound, we show that, even when both the timescale of the rare event and the relative accuracy of the MC estimator are exponentially large in the number of states, LSTD maintains a fixed level of relative accuracy with a total number of observed transitions of the Markov chain that is only \emph{polynomially} large in the number of states.
comment: Final camera-ready version published at NeurIPS 2024. Correct an assumption statement and typos, and change/add a few sentences from the last version
♻ ☆ Model-Based Transfer Learning for Contextual Reinforcement Learning NeurIPS 2024
Deep reinforcement learning (RL) is a powerful approach to complex decision making. However, one issue that limits its practical application is its brittleness, sometimes failing to train in the presence of small changes in the environment. Motivated by the success of zero-shot transfer-where pre-trained models perform well on related tasks-we consider the problem of selecting a good set of training tasks to maximize generalization performance across a range of tasks. Given the high cost of training, it is critical to select training tasks strategically, but not well understood how to do so. We hence introduce Model-Based Transfer Learning (MBTL), which layers on top of existing RL methods to effectively solve contextual RL problems. MBTL models the generalization performance in two parts: 1) the performance set point, modeled using Gaussian processes, and 2) performance loss (generalization gap), modeled as a linear function of contextual similarity. MBTL combines these two pieces of information within a Bayesian optimization (BO) framework to strategically select training tasks. We show theoretically that the method exhibits sublinear regret in the number of training tasks and discuss conditions to further tighten regret bounds. We experimentally validate our methods using urban traffic and standard continuous control benchmarks. The experimental results suggest that MBTL can achieve up to 43x improved sample efficiency compared with canonical independent training and multi-task training. Further experiments demonstrate the efficacy of BO and the insensitivity to the underlying RL algorithm and hyperparameters. This work lays the foundations for investigating explicit modeling of generalization, thereby enabling principled yet effective methods for contextual RL.
comment: 38th Conference on Neural Information Processing Systems (NeurIPS 2024)
♻ ☆ Statistical Efficiency of Distributional Temporal Difference Learning and Freedman's Inequality in Hilbert Spaces
Distributional reinforcement learning (DRL) has achieved empirical success in various domains. One core task in DRL is distributional policy evaluation, which involves estimating the return distribution $\eta^\pi$ for a given policy $\pi$. Distributional temporal difference learning has been accordingly proposed, which extends the classic temporal difference learning (TD) in RL. In this paper, we focus on the non-asymptotic statistical rates of distributional TD. To facilitate theoretical analysis, we propose non-parametric distributional TD (NTD). For a $\gamma$-discounted infinite-horizon tabular Markov decision process, we show that for NTD with a generative model, we need $\tilde{O}(\varepsilon^{-2}\mu_{\min}^{-1}(1-\gamma)^{-3})$ interactions with the environment to achieve an $\varepsilon$-optimal estimator with high probability, when the estimation error is measured by the $1$-Wasserstein. This sample complexity bound is minimax optimal up to logarithmic factors. In addition, we revisit categorical distributional TD (CTD), showing that the same non-asymptotic convergence bounds hold for CTD in the case of the $1$-Wasserstein distance. We also extend our analysis to the more general setting where the data generating process is Markovian. In the Markovian setting, we propose variance-reduced variants of NTD and CTD, and show that both can achieve a $\tilde{O}(\varepsilon^{-2} \mu_{\pi,\min}^{-1}(1-\gamma)^{-3}+t_{mix}\mu_{\pi,\min}^{-1}(1-\gamma)^{-1})$ sample complexity bounds in the case of the $1$-Wasserstein distance, which matches the state-of-the-art statistical results for classic policy evaluation. To achieve the sharp statistical rates, we establish a novel Freedman's inequality in Hilbert spaces. This new Freedman's inequality would be of independent interest for statistical analysis of various infinite-dimensional online learning problems.
♻ ☆ A General Framework for Inference-time Scaling and Steering of Diffusion Models
Diffusion models produce impressive results in modalities ranging from images and video to protein design and text. However, generating samples with user-specified properties remains a challenge. Recent research proposes fine-tuning models to maximize rewards that capture desired properties, but these methods require expensive training and are prone to mode collapse. In this work, we propose Feynman Kac (FK) steering, an inference-time framework for steering diffusion models with reward functions. FK steering works by sampling a system of multiple interacting diffusion processes, called particles, and resampling particles at intermediate steps based on scores computed using functions called potentials. Potentials are defined using rewards for intermediate states and are selected such that a high value indicates that the particle will yield a high-reward sample. We explore various choices of potentials, intermediate rewards, and samplers. We evaluate FK steering on text-to-image and text diffusion models. For steering text-to-image models with a human preference reward, we find that FK steering a 0.8B parameter model outperforms a 2.6B parameter fine-tuned model on prompt fidelity, with faster sampling and no training. For steering text diffusion models with rewards for text quality and specific text attributes, we find that FK steering generates lower perplexity, more linguistically acceptable outputs and enables gradient-free control of attributes like toxicity. Our results demonstrate that inference-time scaling and steering of diffusion models, even with off-the-shelf rewards, can provide significant sample quality gains and controllability benefits. Code is available at https://github.com/zacharyhorvitz/Fk-Diffusion-Steering .
♻ ☆ Smoothness Really Matters: A Simple Yet Effective Approach for Unsupervised Graph Domain Adaptation AAAI2025
Unsupervised Graph Domain Adaptation (UGDA) seeks to bridge distribution shifts between domains by transferring knowledge from labeled source graphs to given unlabeled target graphs. Existing UGDA methods primarily focus on aligning features in the latent space learned by graph neural networks (GNNs) across domains, often overlooking structural shifts, resulting in limited effectiveness when addressing structurally complex transfer scenarios. Given the sensitivity of GNNs to local structural features, even slight discrepancies between source and target graphs could lead to significant shifts in node embeddings, thereby reducing the effectiveness of knowledge transfer. To address this issue, we introduce a novel approach for UGDA called Target-Domain Structural Smoothing (TDSS). TDSS is a simple and effective method designed to perform structural smoothing directly on the target graph, thereby mitigating structural distribution shifts and ensuring the consistency of node representations. Specifically, by integrating smoothing techniques with neighborhood sampling, TDSS maintains the structural coherence of the target graph while mitigating the risk of over-smoothing. Our theoretical analysis shows that TDSS effectively reduces target risk by improving model smoothness. Empirical results on three real-world datasets demonstrate that TDSS outperforms recent state-of-the-art baselines, achieving significant improvements across six transfer scenarios. The code is available in https://github.com/cwei01/TDSS.
comment: 11 pages, Accpected by AAAI2025
♻ ☆ Surveying Attitudinal Alignment Between Large Language Models Vs. Humans Towards 17 Sustainable Development Goals
Large Language Models (LLMs) have emerged as potent tools for advancing the United Nations' Sustainable Development Goals (SDGs). However, the attitudinal disparities between LLMs and humans towards these goals can pose significant challenges. This study conducts a comprehensive review and analysis of the existing literature on the attitudes of LLMs towards the 17 SDGs, emphasizing the comparison between their attitudes and support for each goal and those of humans. We examine the potential disparities, primarily focusing on aspects such as understanding and emotions, cultural and regional differences, task objective variations, and factors considered in the decision-making process. These disparities arise from the underrepresentation and imbalance in LLM training data, historical biases, quality issues, lack of contextual understanding, and skewed ethical values reflected. The study also investigates the risks and harms that may arise from neglecting the attitudes of LLMs towards the SDGs, including the exacerbation of social inequalities, racial discrimination, environmental destruction, and resource wastage. To address these challenges, we propose strategies and recommendations to guide and regulate the application of LLMs, ensuring their alignment with the principles and goals of the SDGs, and therefore creating a more just, inclusive, and sustainable future.
♻ ☆ Autonomous Algorithm for Training Autonomous Vehicles with Minimal Human Intervention
Recent reinforcement learning (RL) algorithms have demonstrated impressive results in simulated driving environments. However, autonomous vehicles trained in simulation often struggle to work well in the real world due to the fidelity gap between simulated and real-world environments. While directly training real-world autonomous vehicles with RL algorithms is a promising approach to bypass the fidelity gap problem, it presents several challenges. One critical yet often overlooked challenge is the need to reset a driving environment between every episode. This reset process demands significant human intervention, leading to poor training efficiency in the real world. In this paper, we introduce a novel autonomous algorithm that enables off-the-shelf RL algorithms to train autonomous vehicles with minimal human intervention. Our algorithm reduces unnecessary human intervention by aborting episodes to prevent unsafe states and identifying informative initial states for subsequent episodes. The key idea behind identifying informative initial states is to estimate the expected amount of information that can be obtained from under-explored but reachable states. Our algorithm also revisits rule-based autonomous driving algorithms and highlights their benefits in safely returning an autonomous vehicle to initial states. To evaluate how much human intervention is required during training, we implement challenging urban driving tasks that require an autonomous vehicle to reset to initial states on its own. The experimental results show that our autonomous algorithm is task-agnostic and achieves competitive driving performance with much less human intervention than baselines.
comment: 8 pages, 6 figures, 2 tables, conference
♻ ☆ Federated Deep Subspace Clustering
This paper introduces FDSC, a private-protected subspace clustering (SC) approach with federated learning (FC) schema. In each client, there is a deep subspace clustering network accounting for grouping the isolated data, composed of a encode network, a self-expressive layer, and a decode network. FDSC is achieved by uploading the encode network to communicate with other clients in the server. Besides, FDSC is also enhanced by preserving the local neighborhood relationship in each client. With the effects of federated learning and locality preservation, the learned data features from the encoder are boosted so as to enhance the self-expressiveness learning and result in better clustering performance. Experiments test FDSC on public datasets and compare with other clustering methods, demonstrating the effectiveness of FDSC.
comment: 8pages,4 figures, 4 Tables
♻ ☆ Gameplay Filters: Robust Zero-Shot Safety through Adversarial Imagination
Despite the impressive recent advances in learning-based robot control, ensuring robustness to out-of-distribution conditions remains an open challenge. Safety filters can, in principle, keep arbitrary control policies from incurring catastrophic failures by overriding unsafe actions, but existing solutions for complex (e.g., legged) robot dynamics do not span the full motion envelope and instead rely on local, reduced-order models. These filters tend to overly restrict agility and can still fail when perturbed away from nominal conditions. This paper presents the gameplay filter, a new class of predictive safety filter that continually plays out hypothetical matches between its simulation-trained safety strategy and a virtual adversary co-trained to invoke worst-case events and sim-to-real error, and precludes actions that would cause failures down the line. We demonstrate the scalability and robustness of the approach with a first-of-its-kind full-order safety filter for (36-D) quadrupedal dynamics. Physical experiments on two different quadruped platforms demonstrate the superior zero-shot effectiveness of the gameplay filter under large perturbations such as tugging and unmodeled terrain. Experiment videos and open-source software are available online: https://saferobotics.org/research/gameplay-filter
♻ ☆ Enhancing Graph Self-Supervised Learning with Graph Interplay
Graph self-supervised learning (GSSL) has emerged as a compelling framework for extracting informative representations from graph-structured data without extensive reliance on labeled inputs. In this study, we introduce Graph Interplay (GIP), an innovative and versatile approach that significantly enhances the performance equipped with various existing GSSL methods. To this end, GIP advocates direct graph-level communications by introducing random inter-graph edges within standard batches. Against GIP's simplicity, we further theoretically show that \textsc{GIP} essentially performs a principled manifold separation via combining inter-graph message passing and GSSL, bringing about more structured embedding manifolds and thus benefits a series of downstream tasks. Our empirical study demonstrates that GIP surpasses the performance of prevailing GSSL methods across multiple benchmarks by significant margins, highlighting its potential as a breakthrough approach. Besides, GIP can be readily integrated into a series of GSSL methods and consistently offers additional performance gain. This advancement not only amplifies the capability of GSSL but also potentially sets the stage for a novel graph learning paradigm in a broader sense.
comment: Due to potential implicit data leakage in our experimental setup, where the pretraining dataset was ordered by default labels, we withdraw this manuscript for further self-examination and rigorous validation
♻ ☆ CryoBench: Diverse and challenging datasets for the heterogeneity problem in cryo-EM NeurIPS 2024
Cryo-electron microscopy (cryo-EM) is a powerful technique for determining high-resolution 3D biomolecular structures from imaging data. Its unique ability to capture structural variability has spurred the development of heterogeneous reconstruction algorithms that can infer distributions of 3D structures from noisy, unlabeled imaging data. Despite the growing number of advanced methods, progress in the field is hindered by the lack of standardized benchmarks with ground truth information and reliable validation metrics. Here, we introduce CryoBench, a suite of datasets, metrics, and benchmarks for heterogeneous reconstruction in cryo-EM. CryoBench includes five datasets representing different sources of heterogeneity and degrees of difficulty. These include conformational heterogeneity generated from designed motions of antibody complexes or sampled from a molecular dynamics simulation, as well as compositional heterogeneity from mixtures of ribosome assembly states or 100 common complexes present in cells. We then analyze state-of-the-art heterogeneous reconstruction tools, including neural and non-neural methods, assess their sensitivity to noise, and propose new metrics for quantitative evaluation. We hope that CryoBench will be a foundational resource for accelerating algorithmic development and evaluation in the cryo-EM and machine learning communities. Project page: https://cryobench.cs.princeton.edu.
comment: Accepted by NeurIPS 2024 (Spotlight)
♻ ☆ An efficient likelihood-free Bayesian inference method based on sequential neural posterior estimation
Sequential neural posterior estimation (SNPE) techniques have been recently proposed for dealing with simulation-based models with intractable likelihoods. Unlike approximate Bayesian computation, SNPE techniques learn the posterior from sequential simulation using neural network-based conditional density estimators by minimizing a specific loss function. The SNPE method proposed by Lueckmann et al. (2017) used a calibration kernel to boost the sample weights around the observed data, resulting in a concentrated loss function. However, the use of calibration kernels may increase the variances of both the empirical loss and its gradient, making the training inefficient. To improve the stability of SNPE, this paper proposes to use an adaptive calibration kernel and several variance reduction techniques. The proposed method greatly speeds up the process of training and provides a better approximation of the posterior than the original SNPE method and some existing competitors as confirmed by numerical experiments. We also managed to demonstrate the superiority of the proposed method for a high-dimensional model with a real-world dataset.
comment: 28 pages, 9 figures
♻ ☆ On the uncertainty principle of neural networks
In this study, we explore the inherent trade-off between accuracy and robustness in neural networks, drawing an analogy to the uncertainty principle in quantum mechanics. We propose that neural networks are subject to an uncertainty relation, which manifests as a fundamental limitation in their ability to simultaneously achieve high accuracy and robustness against adversarial attacks. Through mathematical proofs and empirical evidence, we demonstrate that this trade-off is a natural consequence of the sharp boundaries formed between different class concepts during training. Our findings reveal that the complementarity principle, a cornerstone of quantum physics, applies to neural networks, imposing fundamental limits on their capabilities in simultaneous learning of conjugate features. Meanwhile, our work suggests that achieving human-level intelligence through a single network architecture or massive datasets alone may be inherently limited. Our work provides new insights into the theoretical foundations of neural network vulnerability and opens up avenues for designing more robust neural network architectures.
comment: 8 pages, 5 figures
♻ ☆ Hidden Markov Neural Networks
We define an evolving in-time Bayesian neural network called a Hidden Markov Neural Network, which addresses the crucial challenge in time-series forecasting and continual learning: striking a balance between adapting to new data and appropriately forgetting outdated information. This is achieved by modelling the weights of a neural network as the hidden states of a Hidden Markov model, with the observed process defined by the available data. A filtering algorithm is employed to learn a variational approximation of the evolving-in-time posterior distribution over the weights. By leveraging a sequential variant of Bayes by Backprop, enriched with a stronger regularization technique called variational DropConnect, Hidden Markov Neural Networks achieve robust regularization and scalable inference. Experiments on MNIST, dynamic classification tasks, and next-frame forecasting in videos demonstrate that Hidden Markov Neural Networks provide strong predictive performance while enabling effective uncertainty quantification.
♻ ☆ The Dark Side of Rich Rewards: Understanding and Mitigating Noise in VLM Rewards
While Vision-Language Models (VLMs) are increasingly used to generate reward signals for training embodied agents to follow instructions, our research reveals that agents guided by VLM rewards often underperform compared to those employing only intrinsic (exploration-driven) rewards, contradicting expectations set by recent work. We hypothesize that false positive rewards -- instances where unintended trajectories are incorrectly rewarded -- are more detrimental than false negatives. Our analysis confirms this hypothesis, revealing that the widely used cosine similarity metric is prone to false positive reward estimates. To address this, we introduce BiMI ({Bi}nary {M}utual {I}nformation), a novel reward function designed to mitigate noise. BiMI significantly enhances learning efficiency across diverse and challenging embodied navigation environments. Our findings offer a nuanced understanding of how different types of reward noise impact agent learning and highlight the importance of addressing multimodal reward signal noise when training embodied agents
comment: 11 main body pages, 21 appendix pages
♻ ☆ A Mechanistic Explanatory Strategy for XAI
Despite significant advancements in XAI, scholars note a persistent lack of solid conceptual foundations and integration with broader scientific discourse on explanation. In response, emerging XAI research draws on explanatory strategies from various sciences and philosophy of science literature to fill these gaps. This paper outlines a mechanistic strategy for explaining the functional organization of deep learning systems, situating recent advancements in AI explainability within a broader philosophical context. According to the mechanistic approach, the explanation of opaque AI systems involves identifying mechanisms that drive decision-making. For deep neural networks, this means discerning functionally relevant components -- such as neurons, layers, circuits, or activation patterns -- and understanding their roles through decomposition, localization, and recomposition. Proof-of-principle case studies from image recognition and language modeling align these theoretical approaches with the latest research from AI labs like OpenAI and Anthropic. This research suggests that a systematic approach to studying model organization can reveal elements that simpler (or ''more modest'') explainability techniques might miss, fostering more thoroughly explainable AI. The paper concludes with a discussion on the epistemic relevance of the mechanistic approach positioned in the context of selected philosophical debates on XAI.
comment: Forthcoming in M\"uller, V. C., Dewey, A. R., Dung, L., & L\"ohr, G. (Eds.), Philosophy of Artificial Intelligence: The State of the Art, Synthese Library, Berlin: Springer Nature. Please cite the published version
♻ ☆ Preference-based Pure Exploration
We study the preference-based pure exploration problem for bandits with vector-valued rewards. The rewards are ordered using a (given) preference cone $\mathcal{C}$ and our goal is to identify the set of Pareto optimal arms. First, to quantify the impact of preferences, we derive a novel lower bound on sample complexity for identifying the most preferred policy with a confidence level $1-\delta$. Our lower bound elicits the role played by the geometry of the preference cone and punctuates the difference in hardness compared to existing best-arm identification variants of the problem. We further explicate this geometry when the rewards follow Gaussian distributions. We then provide a convex relaxation of the lower bound and leverage it to design the Preference-based Track and Stop (PreTS) algorithm that identifies the most preferred policy. Finally, we show that the sample complexity of PreTS is asymptotically tight by deriving a new concentration inequality for vector-valued rewards.
♻ ☆ Intelligent Icing Detection Model of Wind Turbine Blades Based on SCADA data
Diagnosis of ice accretion on wind turbine blades is all the time a hard nut to crack in condition monitoring of wind farms. Existing methods focus on mechanism analysis of icing process, deviation degree analysis of feature engineering. However, there have not been deep researches of neural networks applied in this field at present. Supervisory control and data acquisition (SCADA) makes it possible to train networks through continuously providing not only operation parameters and performance parameters of wind turbines but also environmental parameters and operation modes. This paper explores the possibility that using convolutional neural networks (CNNs), generative adversarial networks (GANs) and domain adaption learning to establish intelligent diagnosis frameworks under different training scenarios. Specifically, PGANC and PGANT are proposed for sufficient and non-sufficient target wind turbine labeled data, respectively. The basic idea is that we consider a two-stage training with parallel GANs, which are aimed at capturing intrinsic features for normal and icing samples, followed by classification CNN or domain adaption module in various training cases. Model validation on three wind turbine SCADA data shows that two-stage training can effectively improve the model performance. Besides, if there is no sufficient labeled data for a target turbine, which is an extremely common phenomenon in real industrial practices, the addition of domain adaption learning makes the trained model show better performance. Overall, our proposed intelligent diagnosis frameworks can achieve more accurate detection on the same wind turbine and more generalized capability on a new wind turbine, compared with other machine learning models and conventional CNNs.
comment: 10 pages, 6 figures
♻ ☆ Multi-hop Upstream Anticipatory Traffic Signal Control with Deep Reinforcement Learning
Coordination in traffic signal control is crucial for managing congestion in urban networks. Existing pressure-based control methods focus only on immediate upstream links, leading to suboptimal green time allocation and increased network delays. However, effective signal control inherently requires coordination across a broader spatial scope, as the effect of upstream traffic should influence signal control decisions at downstream intersections, impacting a large area in the traffic network. Although agent communication using neural network-based feature extraction can implicitly enhance spatial awareness, it significantly increases the learning complexity, adding an additional layer of difficulty to the challenging task of control in deep reinforcement learning. To address the issue of learning complexity and myopic traffic pressure definition, our work introduces a novel concept based on Markov chain theory, namely \textit{multi-hop upstream pressure}, which generalizes the conventional pressure to account for traffic conditions beyond the immediate upstream links. This farsighted and compact metric informs the deep reinforcement learning agent to preemptively clear the multi-hop upstream queues, guiding the agent to optimize signal timings with a broader spatial awareness. Simulations on synthetic and realistic (Toronto) scenarios demonstrate controllers utilizing multi-hop upstream pressure significantly reduce overall network delay by prioritizing traffic movements based on a broader understanding of upstream congestion.
comment: 5 tables, 11 figures
♻ ☆ Model Alignment Search
When can we say that two neural systems are the same? The answer to this question is goal-dependent, and it is often addressed through correlative methods such as Representational Similarity Analysis (RSA) and Centered Kernel Alignment (CKA). What do we miss when we forgo causal explorations, and how can we target specific types of similarity? In this work, we introduce Model Alignment Search (MAS), a method for causally exploring distributed representational similarity. The method learns invertible linear transformations that align a subspace between two distributed networks' representations where causal information can be freely interchanged. We first show that the method can be used to transfer specific causal variables, such as the number of items in a counting task, between networks with different training seeds. We then explore open questions in number cognition by comparing different types of numeric representations in models trained on structurally different numeric tasks. We then explore differences between MAS vs preexisting causal similarity methods, and lastly, we introduce a counterfactual latent auxiliary loss function that helps shape causally relevant alignments even in cases where we do not have causal access to one of the two models for training.
♻ ☆ Reinforcement learning with non-ergodic reward increments: robustness via ergodicity transformations
Envisioned application areas for reinforcement learning (RL) include autonomous driving, precision agriculture, and finance, which all require RL agents to make decisions in the real world. A significant challenge hindering the adoption of RL methods in these domains is the non-robustness of conventional algorithms. In particular, the focus of RL is typically on the expected value of the return. The expected value is the average over the statistical ensemble of infinitely many trajectories, which can be uninformative about the performance of the average individual. For instance, when we have a heavy-tailed return distribution, the ensemble average can be dominated by rare extreme events. Consequently, optimizing the expected value can lead to policies that yield exceptionally high returns with a probability that approaches zero but almost surely result in catastrophic outcomes in single long trajectories. In this paper, we develop an algorithm that lets RL agents optimize the long-term performance of individual trajectories. The algorithm enables the agents to learn robust policies, which we show in an instructive example with a heavy-tailed return distribution and standard RL benchmarks. The key element of the algorithm is a transformation that we learn from data. This transformation turns the time series of collected returns into one for whose increments expected value and the average over a long trajectory coincide. Optimizing these increments results in robust policies.
comment: Accepted final version to appear in the Transactions on Machine Learning Research
Artificial Intelligence 116
☆ Learnings from Scaling Visual Tokenizers for Reconstruction and Generation
Visual tokenization via auto-encoding empowers state-of-the-art image and video generative models by compressing pixels into a latent space. Although scaling Transformer-based generators has been central to recent advances, the tokenizer component itself is rarely scaled, leaving open questions about how auto-encoder design choices influence both its objective of reconstruction and downstream generative performance. Our work aims to conduct an exploration of scaling in auto-encoders to fill in this blank. To facilitate this exploration, we replace the typical convolutional backbone with an enhanced Vision Transformer architecture for Tokenization (ViTok). We train ViTok on large-scale image and video datasets far exceeding ImageNet-1K, removing data constraints on tokenizer scaling. We first study how scaling the auto-encoder bottleneck affects both reconstruction and generation -- and find that while it is highly correlated with reconstruction, its relationship with generation is more complex. We next explored the effect of separately scaling the auto-encoders' encoder and decoder on reconstruction and generation performance. Crucially, we find that scaling the encoder yields minimal gains for either reconstruction or generation, while scaling the decoder boosts reconstruction but the benefits for generation are mixed. Building on our exploration, we design ViTok as a lightweight auto-encoder that achieves competitive performance with state-of-the-art auto-encoders on ImageNet-1K and COCO reconstruction tasks (256p and 512p) while outperforming existing auto-encoders on 16-frame 128p video reconstruction for UCF-101, all with 2-5x fewer FLOPs. When integrated with Diffusion Transformers, ViTok demonstrates competitive performance on image generation for ImageNet-1K and sets new state-of-the-art benchmarks for class-conditional video generation on UCF-101.
comment: 28 pages, 25 figures, 7 Tables
☆ KU AIGEN ICL EDI@BC8 Track 3: Advancing Phenotype Named Entity Recognition and Normalization for Dysmorphology Physical Examination Reports
The objective of BioCreative8 Track 3 is to extract phenotypic key medical findings embedded within EHR texts and subsequently normalize these findings to their Human Phenotype Ontology (HPO) terms. However, the presence of diverse surface forms in phenotypic findings makes it challenging to accurately normalize them to the correct HPO terms. To address this challenge, we explored various models for named entity recognition and implemented data augmentation techniques such as synonym marginalization to enhance the normalization step. Our pipeline resulted in an exact extraction and normalization F1 score 2.6\% higher than the mean score of all submissions received in response to the challenge. Furthermore, in terms of the normalization F1 score, our approach surpassed the average performance by 1.9\%. These findings contribute to the advancement of automated medical data extraction and normalization techniques, showcasing potential pathways for future research and application in the biomedical domain.
comment: This article is part of the Proceedings of the BioCreative VIII Challenge and Workshop: Curation and Evaluation in the era of Generative Models
☆ Parallel multi-objective metaheuristics for smart communications in vehicular networks
This article analyzes the use of two parallel multi-objective soft computing algorithms to automatically search for high-quality settings of the Ad hoc On Demand Vector routing protocol for vehicular networks. These methods are based on an evolutionary algorithm and on a swarm intelligence approach. The experimental analysis demonstrates that the configurations computed by our optimization algorithms outperform other state-of-the-art optimized ones. In turn, the computational efficiency achieved by all the parallel versions is greater than 87 %. Therefore, the line of work presented in this article represents an efficient framework to improve vehicular communications.
☆ A Simple Aerial Detection Baseline of Multimodal Language Models
The multimodal language models (MLMs) based on generative pre-trained Transformer are considered powerful candidates for unifying various domains and tasks. MLMs developed for remote sensing (RS) have demonstrated outstanding performance in multiple tasks, such as visual question answering and visual grounding. In addition to visual grounding that detects specific objects corresponded to given instruction, aerial detection, which detects all objects of multiple categories, is also a valuable and challenging task for RS foundation models. However, aerial detection has not been explored by existing RS MLMs because the autoregressive prediction mechanism of MLMs differs significantly from the detection outputs. In this paper, we present a simple baseline for applying MLMs to aerial detection for the first time, named LMMRotate. Specifically, we first introduce a normalization method to transform detection outputs into textual outputs to be compatible with the MLM framework. Then, we propose a evaluation method, which ensures a fair comparison between MLMs and conventional object detection models. We construct the baseline by fine-tuning open-source general-purpose MLMs and achieve impressive detection performance comparable to conventional detector. We hope that this baseline will serve as a reference for future MLM development, enabling more comprehensive capabilities for understanding RS images. Code is available at https://github.com/Li-Qingyun/mllm-mmrotate.
comment: 4 pages, 1 table, 4 figures
☆ CyberMentor: AI Powered Learning Tool Platform to Address Diverse Student Needs in Cybersecurity Education
Many non-traditional students in cybersecurity programs often lack access to advice from peers, family members and professors, which can hinder their educational experiences. Additionally, these students may not fully benefit from various LLM-powered AI assistants due to issues like content relevance, locality of advice, minimum expertise, and timing. This paper addresses these challenges by introducing an application designed to provide comprehensive support by answering questions related to knowledge, skills, and career preparation advice tailored to the needs of these students. We developed a learning tool platform, CyberMentor, to address the diverse needs and pain points of students majoring in cybersecurity. Powered by agentic workflow and Generative Large Language Models (LLMs), the platform leverages Retrieval-Augmented Generation (RAG) for accurate and contextually relevant information retrieval to achieve accessibility and personalization. We demonstrated its value in addressing knowledge requirements for cybersecurity education and for career marketability, in tackling skill requirements for analytical and programming assignments, and in delivering real time on demand learning support. Using three use scenarios, we showcased CyberMentor in facilitating knowledge acquisition and career preparation and providing seamless skill-based guidance and support. We also employed the LangChain prompt-based evaluation methodology to evaluate the platform's impact, confirming its strong performance in helpfulness, correctness, and completeness. These results underscore the system's ability to support students in developing practical cybersecurity skills while improving equity and sustainability within higher education. Furthermore, CyberMentor's open-source design allows for adaptation across other disciplines, fostering educational innovation and broadening its potential impact.
comment: 11 pages, 8 figures
☆ The Goofus & Gallant Story Corpus for Practical Value Alignment ICML
Values or principles are key elements of human society that influence people to behave and function according to an accepted standard set of social rules to maintain social order. As AI systems are becoming ubiquitous in human society, it is a major concern that they could violate these norms or values and potentially cause harm. Thus, to prevent intentional or unintentional harm, AI systems are expected to take actions that align with these principles. Training systems to exhibit this type of behavior is difficult and often requires a specialized dataset. This work presents a multi-modal dataset illustrating normative and non-normative behavior in real-life situations described through natural language and artistic images. This training set contains curated sets of images that are designed to teach young children about social principles. We argue that this is an ideal dataset to use for training socially normative agents given this fact.
comment: Accepted by International Conference on Machine Learning and Applications (ICMLA) 2024. Main Conference, Long Paper
☆ Practical Continual Forgetting for Pre-trained Vision Models
For privacy and security concerns, the need to erase unwanted information from pre-trained vision models is becoming evident nowadays. In real-world scenarios, erasure requests originate at any time from both users and model owners, and these requests usually form a sequence. Therefore, under such a setting, selective information is expected to be continuously removed from a pre-trained model while maintaining the rest. We define this problem as continual forgetting and identify three key challenges. (i) For unwanted knowledge, efficient and effective deleting is crucial. (ii) For remaining knowledge, the impact brought by the forgetting procedure should be minimal. (iii) In real-world scenarios, the training samples may be scarce or partially missing during the process of forgetting. To address them, we first propose Group Sparse LoRA (GS-LoRA). Specifically, towards (i), we introduce LoRA modules to fine-tune the FFN layers in Transformer blocks for each forgetting task independently, and towards (ii), a simple group sparse regularization is adopted, enabling automatic selection of specific LoRA groups and zeroing out the others. To further extend GS-LoRA to more practical scenarios, we incorporate prototype information as additional supervision and introduce a more practical approach, GS-LoRA++. For each forgotten class, we move the logits away from its original prototype. For the remaining classes, we pull the logits closer to their respective prototypes. We conduct extensive experiments on face recognition, object detection and image classification and demonstrate that our method manages to forget specific classes with minimal impact on other classes. Codes have been released on https://github.com/bjzhb666/GS-LoRA.
☆ Cueless EEG imagined speech for subject identification: dataset and benchmarks
Electroencephalogram (EEG) signals have emerged as a promising modality for biometric identification. While previous studies have explored the use of imagined speech with semantically meaningful words for subject identification, most have relied on additional visual or auditory cues. In this study, we introduce a cueless EEG-based imagined speech paradigm, where subjects imagine the pronunciation of semantically meaningful words without any external cues. This innovative approach addresses the limitations of prior methods by requiring subjects to select and imagine words from a predefined list naturally. The dataset comprises over 4,350 trials from 11 subjects across five sessions. We assess a variety of classification methods, including traditional machine learning techniques such as Support Vector Machines (SVM) and XGBoost, as well as time-series foundation models and deep learning architectures specifically designed for EEG classification, such as EEG Conformer and Shallow ConvNet. A session-based hold-out validation strategy was employed to ensure reliable evaluation and prevent data leakage. Our results demonstrate outstanding classification accuracy, reaching 97.93%. These findings highlight the potential of cueless EEG paradigms for secure and reliable subject identification in real-world applications, such as brain-computer interfaces (BCIs).
☆ Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models
Language has long been conceived as an essential tool for human reasoning. The breakthrough of Large Language Models (LLMs) has sparked significant research interest in leveraging these models to tackle complex reasoning tasks. Researchers have moved beyond simple autoregressive token generation by introducing the concept of "thought" -- a sequence of tokens representing intermediate steps in the reasoning process. This innovative paradigm enables LLMs' to mimic complex human reasoning processes, such as tree search and reflective thinking. Recently, an emerging trend of learning to reason has applied reinforcement learning (RL) to train LLMs to master reasoning processes. This approach enables the automatic generation of high-quality reasoning trajectories through trial-and-error search algorithms, significantly expanding LLMs' reasoning capacity by providing substantially more training data. Furthermore, recent studies demonstrate that encouraging LLMs to "think" with more tokens during test-time inference can further significantly boost reasoning accuracy. Therefore, the train-time and test-time scaling combined to show a new research frontier -- a path toward Large Reasoning Model. The introduction of OpenAI's o1 series marks a significant milestone in this research direction. In this survey, we present a comprehensive review of recent progress in LLM reasoning. We begin by introducing the foundational background of LLMs and then explore the key technical components driving the development of large reasoning models, with a focus on automated data construction, learning-to-reason techniques, and test-time scaling. We also analyze popular open-source projects at building large reasoning models, and conclude with open challenges and future research directions.
comment: 36 pages, 5 figures
☆ Reward-Guided Controlled Generation for Inference-Time Alignment in Diffusion Models: Tutorial and Review
This tutorial provides an in-depth guide on inference-time guidance and alignment methods for optimizing downstream reward functions in diffusion models. While diffusion models are renowned for their generative modeling capabilities, practical applications in fields such as biology often require sample generation that maximizes specific metrics (e.g., stability, affinity in proteins, closeness to target structures). In these scenarios, diffusion models can be adapted not only to generate realistic samples but also to explicitly maximize desired measures at inference time without fine-tuning. This tutorial explores the foundational aspects of such inference-time algorithms. We review these methods from a unified perspective, demonstrating that current techniques -- such as Sequential Monte Carlo (SMC)-based guidance, value-based sampling, and classifier guidance -- aim to approximate soft optimal denoising processes (a.k.a. policies in RL) that combine pre-trained denoising processes with value functions serving as look-ahead functions that predict from intermediate states to terminal rewards. Within this framework, we present several novel algorithms not yet covered in the literature. Furthermore, we discuss (1) fine-tuning methods combined with inference-time techniques, (2) inference-time algorithms based on search algorithms such as Monte Carlo tree search, which have received limited attention in current research, and (3) connections between inference-time algorithms in language models and diffusion models. The code of this tutorial on protein design is available at https://github.com/masa-ue/AlignInversePro
comment: We plan to add more content/codes. Please let us know if there are any comments
☆ Incorporating Quantum Advantage in Quantum Circuit Generation through Genetic Programming
Designing efficient quantum circuits that leverage quantum advantage compared to classical computing has become increasingly critical. Genetic algorithms have shown potential in generating such circuits through artificial evolution. However, integrating quantum advantage into the fitness function of these algorithms remains unexplored. In this paper, we aim to enhance the efficiency of quantum circuit design by proposing two novel approaches for incorporating quantum advantage metrics into the fitness function of genetic algorithms.1 We evaluate our approaches based on the Bernstein-Vazirani Problem and the Unstructured Database Search Problem as test cases. The results demonstrate that our approaches not only improve the convergence speed of the genetic algorithm but also produce circuits comparable to expert-designed solutions. Our findings suggest that automated quantum circuit design using genetic algorithms that incorporate a measure of quantum advantage is a promising approach to accelerating the development of quantum algorithms.
☆ Authenticated Delegation and Authorized AI Agents
The rapid deployment of autonomous AI agents creates urgent challenges around authorization, accountability, and access control in digital spaces. New standards are needed to know whom AI agents act on behalf of and guide their use appropriately, protecting online spaces while unlocking the value of task delegation to autonomous agents. We introduce a novel framework for authenticated, authorized, and auditable delegation of authority to AI agents, where human users can securely delegate and restrict the permissions and scope of agents while maintaining clear chains of accountability. This framework builds on existing identification and access management protocols, extending OAuth 2.0 and OpenID Connect with agent-specific credentials and metadata, maintaining compatibility with established authentication and web infrastructure. Further, we propose a framework for translating flexible, natural language permissions into auditable access control configurations, enabling robust scoping of AI agent capabilities across diverse interaction modalities. Taken together, this practical approach facilitates immediate deployment of AI agents while addressing key security and accountability concerns, working toward ensuring agentic AI systems perform only appropriate actions and providing a tool for digital service providers to enable AI agent interactions without risking harm from scalable interaction.
☆ Robin: a Suite of Multi-Scale Vision-Language Models and the CHIRP Evaluation Benchmark
The proliferation of Vision-Language Models (VLMs) in the past several years calls for rigorous and comprehensive evaluation methods and benchmarks. This work analyzes existing VLM evaluation techniques, including automated metrics, AI-based assessments, and human evaluations across diverse tasks. We first introduce Robin - a novel suite of VLMs that we built by combining Large Language Models (LLMs) and Vision Encoders (VEs) at multiple scales, and use Robin to identify shortcomings of current evaluation approaches across scales. Next, to overcome the identified limitations, we introduce CHIRP - a new long form response benchmark we developed for more robust and complete VLM evaluation. We provide open access to the Robin training code, model suite, and CHIRP benchmark to promote reproducibility and advance VLM research.
☆ The Heap: A Contamination-Free Multilingual Code Dataset for Evaluating Large Language Models
The recent rise in the popularity of large language models has spurred the development of extensive code datasets needed to train them. This has left limited code available for collection and use in the downstream investigation of specific behaviors, or evaluation of large language models without suffering from data contamination. To address this problem, we release The Heap, a large multilingual dataset covering 57 programming languages that has been deduplicated with respect to other open datasets of code, enabling researchers to conduct fair evaluations of large language models without significant data cleaning overhead.
comment: Pre-Print. Accepted to FORGE 2025 Dataset Track
☆ Monte Carlo Tree Search with Velocity Obstacles for safe and efficient motion planning in dynamic environments
Online motion planning is a challenging problem for intelligent robots moving in dense environments with dynamic obstacles, e.g., crowds. In this work, we propose a novel approach for optimal and safe online motion planning with minimal information about dynamic obstacles. Specifically, our approach requires only the current position of the obstacles and their maximum speed, but it does not need any information about their exact trajectories or dynamic model. The proposed methodology combines Monte Carlo Tree Search (MCTS), for online optimal planning via model simulations, with Velocity Obstacles (VO), for obstacle avoidance. We perform experiments in a cluttered simulated environment with walls, and up to 40 dynamic obstacles moving with random velocities and directions. With an ablation study, we show the key contribution of VO in scaling up the efficiency of MCTS, selecting the safest and most rewarding actions in the tree of simulations. Moreover, we show the superiority of our methodology with respect to state-of-the-art planners, including Non-linear Model Predictive Control (NMPC), in terms of improved collision rate, computational and task performance.
☆ NS-Gym: Open-Source Simulation Environments and Benchmarks for Non-Stationary Markov Decision Processes
In many real-world applications, agents must make sequential decisions in environments where conditions are subject to change due to various exogenous factors. These non-stationary environments pose significant challenges to traditional decision-making models, which typically assume stationary dynamics. Non-stationary Markov decision processes (NS-MDPs) offer a framework to model and solve decision problems under such changing conditions. However, the lack of standardized benchmarks and simulation tools has hindered systematic evaluation and advance in this field. We present NS-Gym, the first simulation toolkit designed explicitly for NS-MDPs, integrated within the popular Gymnasium framework. In NS-Gym, we segregate the evolution of the environmental parameters that characterize non-stationarity from the agent's decision-making module, allowing for modular and flexible adaptations to dynamic environments. We review prior work in this domain and present a toolkit encapsulating key problem characteristics and types in NS-MDPs. This toolkit is the first effort to develop a set of standardized interfaces and benchmark problems to enable consistent and reproducible evaluation of algorithms under non-stationary conditions. We also benchmark six algorithmic approaches from prior work on NS-MDPs using NS-Gym. Our vision is that NS-Gym will enable researchers to assess the adaptability and robustness of their decision-making algorithms to non-stationary conditions.
comment: 23 pages, 17 figures
☆ CarMem: Enhancing Long-Term Memory in LLM Voice Assistants through Category-Bounding COLING 2025
In today's assistant landscape, personalisation enhances interactions, fosters long-term relationships, and deepens engagement. However, many systems struggle with retaining user preferences, leading to repetitive user requests and disengagement. Furthermore, the unregulated and opaque extraction of user preferences in industry applications raises significant concerns about privacy and trust, especially in regions with stringent regulations like Europe. In response to these challenges, we propose a long-term memory system for voice assistants, structured around predefined categories. This approach leverages Large Language Models to efficiently extract, store, and retrieve preferences within these categories, ensuring both personalisation and transparency. We also introduce a synthetic multi-turn, multi-session conversation dataset (CarMem), grounded in real industry data, tailored to an in-car voice assistant setting. Benchmarked on the dataset, our system achieves an F1-score of .78 to .95 in preference extraction, depending on category granularity. Our maintenance strategy reduces redundant preferences by 95% and contradictory ones by 92%, while the accuracy of optimal retrieval is at .87. Collectively, the results demonstrate the system's suitability for industrial applications.
comment: Accepted for presentation at the International Conference on Computational Linguistics (COLING 2025)
☆ Electronic Health Records: Towards Digital Twins in Healthcare
The pivotal shift from traditional paper-based records to sophisticated Electronic Health Records (EHR), enabled systematic collection and analysis of patient data through descriptive statistics, providing insight into patterns and trends across patient populations. This evolution continued toward predictive analytics, allowing healthcare providers to anticipate patient outcomes and potential complications before they occur. This progression from basic digital record-keeping to sophisticated predictive modelling and digital twins reflects healthcare's broader evolution toward more integrated, patient-centred approaches that combine data-driven insights with personalized care delivery. This chapter explores the evolution and significance of healthcare information systems, beginning with an examination of the implementation of EHR in the UK and the USA. It provides a comprehensive overview of the International Classification of Diseases (ICD) system, tracing its development from ICD-9 to ICD-10. Central to this discussion is the MIMIC-III database, a landmark achievement in healthcare data sharing and arguably the most comprehensive critical care database freely available to researchers worldwide. MIMIC-III has democratized access to high-quality healthcare data, enabling unprecedented opportunities for research and analysis. The chapter examines its structure, clinical outcome analysis capabilities, and practical applications through case studies, with a particular focus on mortality and length of stay metrics, vital signs extraction, and ICD coding. Through detailed entity-relationship diagrams and practical examples, the text illustrates MIMIC's complex data structure and demonstrates how different querying approaches can lead to subtly different results, emphasizing the critical importance of understanding the database's architecture for accurate data extraction.
☆ Platform-Aware Mission Planning
Planning for autonomous systems typically requires reasoning with models at different levels of abstraction, and the harmonization of two competing sets of objectives: high-level mission goals that refer to an interaction of the system with the external environment, and low-level platform constraints that aim to preserve the integrity and the correct interaction of the subsystems. The complicated interplay between these two models makes it very hard to reason on the system as a whole, especially when the objective is to find plans with robustness guarantees, considering the non-deterministic behavior of the lower layers of the system. In this paper, we introduce the problem of Platform-Aware Mission Planning (PAMP), addressing it in the setting of temporal durative actions. The PAMP problem differs from standard temporal planning for its exists-forall nature: the high-level plan dealing with mission goals is required to satisfy safety and executability constraints, for all the possible non-deterministic executions of the low-level model of the platform and the environment. We propose two approaches for solving PAMP. The first baseline approach amalgamates the mission and platform levels, while the second is based on an abstraction-refinement loop that leverages the combination of a planner and a verification engine. We prove the soundness and completeness of the proposed approaches and validate them experimentally, demonstrating the importance of heterogeneous modeling and the superiority of the technique based on abstraction-refinement.
☆ Artificial Intelligence-Driven Clinical Decision Support Systems
As artificial intelligence (AI) becomes increasingly embedded in healthcare delivery, this chapter explores the critical aspects of developing reliable and ethical Clinical Decision Support Systems (CDSS). Beginning with the fundamental transition from traditional statistical models to sophisticated machine learning approaches, this work examines rigorous validation strategies and performance assessment methods, including the crucial role of model calibration and decision curve analysis. The chapter emphasizes that creating trustworthy AI systems in healthcare requires more than just technical accuracy; it demands careful consideration of fairness, explainability, and privacy. The challenge of ensuring equitable healthcare delivery through AI is stressed, discussing methods to identify and mitigate bias in clinical predictive models. The chapter then delves into explainability as a cornerstone of human-centered CDSS. This focus reflects the understanding that healthcare professionals must not only trust AI recommendations but also comprehend their underlying reasoning. The discussion advances in an analysis of privacy vulnerabilities in medical AI systems, from data leakage in deep learning models to sophisticated attacks against model explanations. The text explores privacy-preservation strategies such as differential privacy and federated learning, while acknowledging the inherent trade-offs between privacy protection and model performance. This progression, from technical validation to ethical considerations, reflects the multifaceted challenges of developing AI systems that can be seamlessly and reliably integrated into daily clinical practice while maintaining the highest standards of patient care and data protection.
☆ Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment
Recent advances in large language models (LLMs) have demonstrated significant progress in performing complex tasks. While Reinforcement Learning from Human Feedback (RLHF) has been effective in aligning LLMs with human preferences, it is susceptible to spurious correlations in reward modeling. Consequently, it often introduces biases-such as length bias, sycophancy, conceptual bias, and discrimination that hinder the model's ability to capture true causal relationships. To address this, we propose a novel causal reward modeling approach that integrates causal inference to mitigate these spurious correlations. Our method enforces counterfactual invariance, ensuring reward predictions remain consistent when irrelevant variables are altered. Through experiments on both synthetic and real-world datasets, we show that our approach mitigates various types of spurious correlations effectively, resulting in more reliable and fair alignment of LLMs with human preferences. As a drop-in enhancement to the existing RLHF workflow, our causal reward modeling provides a practical way to improve the trustworthiness and fairness of LLM finetuning.
☆ Metric Learning with Progressive Self-Distillation for Audio-Visual Embedding Learning ICASSP 2025
Metric learning projects samples into an embedded space, where similarities and dissimilarities are quantified based on their learned representations. However, existing methods often rely on label-guided representation learning, where representations of different modalities, such as audio and visual data, are aligned based on annotated labels. This approach tends to underutilize latent complex features and potential relationships inherent in the distributions of audio and visual data that are not directly tied to the labels, resulting in suboptimal performance in audio-visual embedding learning. To address this issue, we propose a novel architecture that integrates cross-modal triplet loss with progressive self-distillation. Our method enhances representation learning by leveraging inherent distributions and dynamically refining soft audio-visual alignments -- probabilistic alignments between audio and visual data that capture the inherent relationships beyond explicit labels. Specifically, the model distills audio-visual distribution-based knowledge from annotated labels in a subset of each batch. This self-distilled knowledge is used t
comment: 5 pages, 3 figures, 2 tables. Accepted by ICASSP 2025
☆ Managed-Retention Memory: A New Class of Memory for the AI Era
AI clusters today are one of the major uses of High Bandwidth Memory (HBM). However, HBM is suboptimal for AI workloads for several reasons. Analysis shows HBM is overprovisioned on write performance, but underprovisioned on density and read bandwidth, and also has significant energy per bit overheads. It is also expensive, with lower yield than DRAM due to manufacturing complexity. We propose a new memory class: Managed-Retention Memory (MRM), which is more optimized to store key data structures for AI inference workloads. We believe that MRM may finally provide a path to viability for technologies that were originally proposed to support Storage Class Memory (SCM). These technologies traditionally offered long-term persistence (10+ years) but provided poor IO performance and/or endurance. MRM makes different trade-offs, and by understanding the workload IO patterns, MRM foregoes long-term data retention and write performance for better potential performance on the metrics important for these workloads.
comment: 8 pages (5 content + 3 refs); 1 figure
☆ Reducing the Sensitivity of Neural Physics Simulators to Mesh Topology via Pretraining
Meshes are used to represent complex objects in high fidelity physics simulators across a variety of domains, such as radar sensing and aerodynamics. There is growing interest in using neural networks to accelerate physics simulations, and also a growing body of work on applying neural networks directly to irregular mesh data. Since multiple mesh topologies can represent the same object, mesh augmentation is typically required to handle topological variation when training neural networks. Due to the sensitivity of physics simulators to small changes in mesh shape, it is challenging to use these augmentations when training neural network-based physics simulators. In this work, we show that variations in mesh topology can significantly reduce the performance of neural network simulators. We evaluate whether pretraining can be used to address this issue, and find that employing an established autoencoder pretraining technique with graph embedding models reduces the sensitivity of neural network simulators to variations in mesh topology. Finally, we highlight future research directions that may further reduce neural simulator sensitivity to mesh topology.
comment: 5 pages, 3 figures
☆ IFRA: a machine learning-based Instrumented Fall Risk Assessment Scale derived from Instrumented Timed Up and Go test in stroke patients
Effective fall risk assessment is critical for post-stroke patients. The present study proposes a novel, data-informed fall risk assessment method based on the instrumented Timed Up and Go (ITUG) test data, bringing in many mobility measures that traditional clinical scales fail to capture. IFRA, which stands for Instrumented Fall Risk Assessment, has been developed using a two-step process: first, features with the highest predictive power among those collected in a ITUG test have been identified using machine learning techniques; then, a strategy is proposed to stratify patients into low, medium, or high-risk strata. The dataset used in our analysis consists of 142 participants, out of which 93 were used for training (15 synthetically generated), 17 for validation and 32 to test the resulting IFRA scale (22 non-fallers and 10 fallers). Features considered in the IFRA scale include gait speed, vertical acceleration during sit-to-walk transition, and turning angular velocity, which align well with established literature on the risk of fall in neurological patients. In a comparison with traditional clinical scales such as the traditional Timed Up & Go and the Mini-BESTest, IFRA demonstrates competitive performance, being the only scale to correctly assign more than half of the fallers to the high-risk stratum (Fischer's Exact test p = 0.004). Despite the dataset's limited size, this is the first proof-of-concept study to pave the way for future evidence regarding the use of IFRA tool for continuous patient monitoring and fall prevention both in clinical stroke rehabilitation and at home post-discharge.
comment: 26 pages, 2 figures, submitted for review dec 2024
☆ MatrixNet: Learning over symmetry groups using learned group representations NeurIPS 2024
Group theory has been used in machine learning to provide a theoretically grounded approach for incorporating known symmetry transformations in tasks from robotics to protein modeling. In these applications, equivariant neural networks use known symmetry groups with predefined representations to learn over geometric input data. We propose MatrixNet, a neural network architecture that learns matrix representations of group element inputs instead of using predefined representations. MatrixNet achieves higher sample efficiency and generalization over several standard baselines in prediction tasks over the several finite groups and the Artin braid group. We also show that MatrixNet respects group relations allowing generalization to group elements of greater word length than in the training set.
comment: NeurIPS 2024
☆ Text-driven Adaptation of Foundation Models for Few-shot Surgical Workflow Analysis
Purpose: Surgical workflow analysis is crucial for improving surgical efficiency and safety. However, previous studies rely heavily on large-scale annotated datasets, posing challenges in cost, scalability, and reliance on expert annotations. To address this, we propose Surg-FTDA (Few-shot Text-driven Adaptation), designed to handle various surgical workflow analysis tasks with minimal paired image-label data. Methods: Our approach has two key components. First, Few-shot selection-based modality alignment selects a small subset of images and aligns their embeddings with text embeddings from the downstream task, bridging the modality gap. Second, Text-driven adaptation leverages only text data to train a decoder, eliminating the need for paired image-text data. This decoder is then applied to aligned image embeddings, enabling image-related tasks without explicit image-text pairs. Results: We evaluate our approach to generative tasks (image captioning) and discriminative tasks (triplet recognition and phase recognition). Results show that Surg-FTDA outperforms baselines and generalizes well across downstream tasks. Conclusion: We propose a text-driven adaptation approach that mitigates the modality gap and handles multiple downstream tasks in surgical workflow analysis, with minimal reliance on large annotated datasets. The code and dataset will be released in https://github.com/TingxuanSix/Surg-FTDA.
☆ AI in Support of Diversity and Inclusion
In this paper, we elaborate on how AI can support diversity and inclusion and exemplify research projects conducted in that direction. We start by looking at the challenges and progress in making large language models (LLMs) more transparent, inclusive, and aware of social biases. Even though LLMs like ChatGPT have impressive abilities, they struggle to understand different cultural contexts and engage in meaningful, human like conversations. A key issue is that biases in language processing, especially in machine translation, can reinforce inequality. Tackling these biases requires a multidisciplinary approach to ensure AI promotes diversity, fairness, and inclusion. We also highlight AI's role in identifying biased content in media, which is important for improving representation. By detecting unequal portrayals of social groups, AI can help challenge stereotypes and create more inclusive technologies. Transparent AI algorithms, which clearly explain their decisions, are essential for building trust and reducing bias in AI systems. We also stress AI systems need diverse and inclusive training data. Projects like the Child Growth Monitor show how using a wide range of data can help address real world problems like malnutrition and poverty. We present a project that demonstrates how AI can be applied to monitor the role of search engines in spreading disinformation about the LGBTQ+ community. Moreover, we discuss the SignON project as an example of how technology can bridge communication gaps between hearing and deaf people, emphasizing the importance of collaboration and mutual trust in developing inclusive AI. Overall, with this paper, we advocate for AI systems that are not only effective but also socially responsible, promoting fair and inclusive interactions between humans and machines.
comment: 14 pages, 2 figures
☆ Class Incremental Fault Diagnosis under Limited Fault Data via Supervised Contrastive Knowledge Distillation
Class-incremental fault diagnosis requires a model to adapt to new fault classes while retaining previous knowledge. However, limited research exists for imbalanced and long-tailed data. Extracting discriminative features from few-shot fault data is challenging, and adding new fault classes often demands costly model retraining. Moreover, incremental training of existing methods risks catastrophic forgetting, and severe class imbalance can bias the model's decisions toward normal classes. To tackle these issues, we introduce a Supervised Contrastive knowledge distiLlation for class Incremental Fault Diagnosis (SCLIFD) framework proposing supervised contrastive knowledge distillation for improved representation learning capability and less forgetting, a novel prioritized exemplar selection method for sample replay to alleviate catastrophic forgetting, and the Random Forest Classifier to address the class imbalance. Extensive experimentation on simulated and real-world industrial datasets across various imbalance ratios demonstrates the superiority of SCLIFD over existing approaches. Our code can be found at https://github.com/Zhang-Henry/SCLIFD_TII.
☆ MonoSOWA: Scalable monocular 3D Object detector Without human Annotations
Detecting the three-dimensional position and orientation of objects using a single RGB camera is a foundational task in computer vision with many important applications. Traditionally, 3D object detection methods are trained in a fully-supervised setup, requiring vast amounts of human annotations, which are laborious, costly, and do not scale well with the ever-increasing amounts of data being captured. In this paper, we present the first method to train 3D object detectors for monocular RGB cameras without domain-specific human annotations, thus making orders of magnitude more data available for training. Thanks to newly proposed Canonical Object Space, the method can not only exploit data across a variety of datasets and camera setups to train a single 3D detector, but unlike previous work it also works out of the box in previously unseen camera setups. All this is crucial for practical applications, where the data and cameras are extremely heterogeneous. The method is evaluated on two standard autonomous driving datasets, where it outperforms previous works, which, unlike our method, still rely on 2D human annotations.
☆ Predicting Air Temperature from Volumetric Urban Morphology with Machine Learning
In this study, we firstly introduce a method that converts CityGML data into voxels which works efficiently and fast in high resolution for large scale datasets such as cities but by sacrificing some building details to overcome the limitations of previous voxelization methodologies that have been computationally intensive and inefficient at transforming large-scale urban areas into voxel representations for high resolution. Those voxelized 3D city data from multiple cities and corresponding air temperature data are used to develop a machine learning model. Before the model training, Gaussian blurring is implemented on input data to consider spatial relationships, as a result the correlation rate between air temperature and volumetric building morphology is also increased after the Gaussian blurring. After the model training, the prediction results are not just evaluated with Mean Square Error (MSE) but some image similarity metrics such as Structural Similarity Index Measure (SSIM) and Learned Perceptual Image Patch Similarity (LPIPS) that are able to detect and consider spatial relations during the evaluation process. This trained model is capable of predicting the spatial distribution of air temperature by using building volume information of corresponding pixel as input. By doing so, this research aims to assist urban planners in incorporating environmental parameters into their planning strategies, thereby facilitating more sustainable and inhabitable urban environments.
comment: 30 pages, 8 figures, 2 tables
☆ RE-POSE: Synergizing Reinforcement Learning-Based Partitioning and Offloading for Edge Object Detection
Object detection plays a crucial role in smart video analysis, with applications ranging from autonomous driving and security to smart cities. However, achieving real-time object detection on edge devices presents significant challenges due to their limited computational resources and the high demands of deep neural network (DNN)-based detection models, particularly when processing high-resolution video. Conventional strategies, such as input down-sampling and network up-scaling, often compromise detection accuracy for faster performance or lead to higher inference latency. To address these issues, this paper introduces RE-POSE, a Reinforcement Learning (RL)-Driven Partitioning and Edge Offloading framework designed to optimize the accuracy-latency trade-off in resource-constrained edge environments. Our approach features an RL-Based Dynamic Clustering Algorithm (RL-DCA) that partitions video frames into non-uniform blocks based on object distribution and the computational characteristics of DNNs. Furthermore, a parallel edge offloading scheme is implemented to distribute these blocks across multiple edge servers for concurrent processing. Experimental evaluations show that RE-POSE significantly enhances detection accuracy and reduces inference latency, surpassing existing methods.
☆ Solving the unsolvable: Translating case law in Hong Kong
This paper addresses the challenges translating case law under Hong Kong's bilingual legal system. It highlights the initial success of translating all written statutes into Chinese before the 1997 handover, a task mandated by the Basic Law. The effort involved significant collaboration among legal, linguistic, and translation experts, resulting in a comprehensive and culturally appropriate bilingual legal system. However, translating case law remains a significant challenge due to the sheer volume and continuous growth of judicial decisions. The paper critiques the governments and judiciarys sporadic and uncoordinated efforts to translate case law, contrasting it with the thorough approach previously taken for statute translation. Although the government acknowledges the importance of legal bilingualism, it lacks a sustainable strategy for translating case law. The Judiciarys position that translating all judgments is unnecessary, unrealistic, and not cost-effectiveis analyzed and critiqued for its impact on legal transparency and public trust. A proposed solution involves leveraging machine translation technology through a human-machine interactive translation platform, which undergoes two major transitions. Initially based on a neural model, the platform transitions to using a large language model for improved translation accuracy. Furthermore, it evolves from a single-agent system to a multi-agent system, incorporating Translator, Annotator, and Proofreader agents. This multi-agent approach, supported by a grant, aims to facilitate efficient, high-quality translation of judicial judgments by integrating advanced artificial intelligence and continuous feedback mechanisms, thus better meeting the needs of a bilingual legal system.
☆ A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy
While large language models (LLMs) present significant potential for supporting numerous real-world applications and delivering positive social impacts, they still face significant challenges in terms of the inherent risk of privacy leakage, hallucinated outputs, and value misalignment, and can be maliciously used for generating toxic content and unethical purposes after been jailbroken. Therefore, in this survey, we present a comprehensive review of recent advancements aimed at mitigating these issues, organized across the four phases of LLM development and usage: data collecting and pre-training, fine-tuning and alignment, prompting and reasoning, and post-processing and auditing. We elaborate on the recent advances for enhancing the performance of LLMs in terms of privacy protection, hallucination reduction, value alignment, toxicity elimination, and jailbreak defenses. In contrast to previous surveys that focus on a single dimension of responsible LLMs, this survey presents a unified framework that encompasses these diverse dimensions, providing a comprehensive view of enhancing LLMs to better serve real-world applications.
☆ ADAGE: A generic two-layer framework for adaptive agent based modelling AAMAS
Agent-based models (ABMs) are valuable for modelling complex, potentially out-of-equilibria scenarios. However, ABMs have long suffered from the Lucas critique, stating that agent behaviour should adapt to environmental changes. Furthermore, the environment itself often adapts to these behavioural changes, creating a complex bi-level adaptation problem. Recent progress integrating multi-agent reinforcement learning into ABMs introduces adaptive agent behaviour, beginning to address the first part of this critique, however, the approaches are still relatively ad hoc, lacking a general formulation, and furthermore, do not tackle the second aspect of simultaneously adapting environmental level characteristics in addition to the agent behaviours. In this work, we develop a generic two-layer framework for ADaptive AGEnt based modelling (ADAGE) for addressing these problems. This framework formalises the bi-level problem as a Stackelberg game with conditional behavioural policies, providing a consolidated framework for adaptive agent-based modelling based on solving a coupled set of non-linear equations. We demonstrate how this generic approach encapsulates several common (previously viewed as distinct) ABM tasks, such as policy design, calibration, scenario generation, and robust behavioural learning under one unified framework. We provide example simulations on multiple complex economic and financial environments, showing the strength of the novel framework under these canonical settings, addressing long-standing critiques of traditional ABMs.
comment: Accepted at the 2025 International Conference on Autonomous Agents and Multiagent Systems (AAMAS)
☆ Dynamic Neural Style Transfer for Artistic Image Generation using VGG19
Throughout history, humans have created remarkable works of art, but artificial intelligence has only recently started to make strides in generating visually compelling art. Breakthroughs in the past few years have focused on using convolutional neural networks (CNNs) to separate and manipulate the content and style of images, applying texture synthesis techniques. Nevertheless, a number of current techniques continue to encounter obstacles, including lengthy processing times, restricted choices of style images, and the inability to modify the weight ratio of styles. We proposed a neural style transfer system that can add various artistic styles to a desired image to address these constraints allowing flexible adjustments to style weight ratios and reducing processing time. The system uses the VGG19 model for feature extraction, ensuring high-quality, flexible stylization without compromising content integrity.
☆ MoE$^2$: Optimizing Collaborative Inference for Edge Large Language Models
Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of natural language processing tasks. Exploiting the heterogeneous capabilities of edge LLMs is crucial for diverse emerging applications, as it enables greater cost-effectiveness and reduced latency. In this work, we introduce \textit{Mixture-of-Edge-Experts (MoE$^2$)}, a novel collaborative inference framework for edge LLMs. We formulate the joint gating and expert selection problem to optimize inference performance under energy and latency constraints. Unlike conventional MoE problems, LLM expert selection is significantly more challenging due to the combinatorial nature and the heterogeneity of edge LLMs across various attributes. To this end, we propose a two-level expert selection mechanism through which we uncover an optimality-preserving property of gating parameters across expert selections. This property enables the decomposition of the training and selection processes, significantly reducing complexity. Furthermore, we leverage the objective's monotonicity and design a discrete monotonic optimization algorithm for optimal expert selection. We implement edge servers with NVIDIA Jetson AGX Orins and NVIDIA RTX 4090 GPUs, and perform extensive experiments. Our results validate that performance improvements of various LLM models and show that our MoE$^2$ method can achieve optimal trade-offs among different delay and energy budgets, and outperforms baselines under various system resource constraints.
comment: Submitted to IEEE/ACM Transactions on Networking
☆ ELM-DeepONets: Backpropagation-Free Training of Deep Operator Networks via Extreme Learning Machines
Deep Operator Networks (DeepONets) are among the most prominent frameworks for operator learning, grounded in the universal approximation theorem for operators. However, training DeepONets typically requires significant computational resources. To address this limitation, we propose ELM-DeepONets, an Extreme Learning Machine (ELM) framework for DeepONets that leverages the backpropagation-free nature of ELM. By reformulating DeepONet training as a least-squares problem for newly introduced parameters, the ELM-DeepONet approach significantly reduces training complexity. Validation on benchmark problems, including nonlinear ODEs and PDEs, demonstrates that the proposed method not only achieves superior accuracy but also drastically reduces computational costs. This work offers a scalable and efficient alternative for operator learning in scientific computing.
☆ Quantum-Enhanced Transformers for Robust Acoustic Scene Classification in IoT Environments
The proliferation of Internet of Things (IoT) devices equipped with acoustic sensors necessitates robust acoustic scene classification (ASC) capabilities, even in noisy and data-limited environments. Traditional machine learning methods often struggle to generalize effectively under such conditions. To address this, we introduce Q-ASC, a novel Quantum-Inspired Acoustic Scene Classifier that leverages the power of quantum-inspired transformers. By integrating quantum concepts like superposition and entanglement, Q-ASC achieves superior feature learning and enhanced noise resilience compared to classical models. Furthermore, we introduce a Quantum Variational Autoencoder (QVAE) based data augmentation technique to mitigate the challenge of limited labeled data in IoT deployments. Extensive evaluations on the Tampere University of Technology (TUT) Acoustic Scenes 2016 benchmark dataset demonstrate that Q-ASC achieves remarkable accuracy between 68.3% and 88.5% under challenging conditions, outperforming state-of-the-art methods by over 5% in the best case. This research paves the way for deploying intelligent acoustic sensing in IoT networks, with potential applications in smart homes, industrial monitoring, and environmental surveillance, even in adverse acoustic environments.
comment: 5 pages, 4 figures
☆ Aligning Instruction Tuning with Pre-training
Instruction tuning enhances large language models (LLMs) to follow human instructions across diverse tasks, relying on high-quality datasets to guide behavior. However, these datasets, whether manually curated or synthetically generated, are often narrowly focused and misaligned with the broad distributions captured during pre-training, limiting LLM generalization and effective use of pre-trained knowledge. We propose *Aligning Instruction Tuning with Pre-training* (AITP), a method that bridges this gap by identifying coverage shortfalls in instruction-tuning datasets and rewriting underrepresented pre-training data into high-quality instruction-response pairs. This approach enriches dataset diversity while preserving task-specific objectives. Evaluations on three fully open LLMs across eight benchmarks demonstrate consistent performance improvements with AITP. Ablations highlight the benefits of adaptive data selection, controlled rewriting, and balanced integration, emphasizing the importance of aligning instruction tuning with pre-training distributions to unlock the full potential of LLMs.
☆ YETI (YET to Intervene) Proactive Interventions by Multimodal AI Agents in Augmented Reality Tasks
Multimodal AI Agents are AI models that have the capability of interactively and cooperatively assisting human users to solve day-to-day tasks. Augmented Reality (AR) head worn devices can uniquely improve the user experience of solving procedural day-to-day tasks by providing egocentric multimodal (audio and video) observational capabilities to AI Agents. Such AR capabilities can help AI Agents see and listen to actions that users take which can relate to multimodal capabilities of human users. Existing AI Agents, either Large Language Models (LLMs) or Multimodal Vision-Language Models (VLMs) are reactive in nature, which means that models cannot take an action without reading or listening to the human user's prompts. Proactivity of AI Agents on the other hand can help the human user detect and correct any mistakes in agent observed tasks, encourage users when they do tasks correctly or simply engage in conversation with the user - akin to a human teaching or assisting a user. Our proposed YET to Intervene (YETI) multimodal agent focuses on the research question of identifying circumstances that may require the agent to intervene proactively. This allows the agent to understand when it can intervene in a conversation with human users that can help the user correct mistakes on tasks, like cooking, using AR. Our YETI Agent learns scene understanding signals based on interpretable notions of Structural Similarity (SSIM) on consecutive video frames. We also define the alignment signal which the AI Agent can learn to identify if the video frames corresponding to the user's actions on the task are consistent with expected actions. These signals are used by our AI Agent to determine when it should proactively intervene. We compare our results on the instances of proactive intervention in the HoloAssist multimodal benchmark for an expert agent guiding a user to complete procedural tasks.
comment: Preprint
☆ Style4Rec: Enhancing Transformer-based E-commerce Recommendation Systems with Style and Shopping Cart Information
Understanding users' product preferences is essential to the efficacy of a recommendation system. Precision marketing leverages users' historical data to discern these preferences and recommends products that align with them. However, recent browsing and purchase records might better reflect current purchasing inclinations. Transformer-based recommendation systems have made strides in sequential recommendation tasks, but they often fall short in utilizing product image style information and shopping cart data effectively. In light of this, we propose Style4Rec, a transformer-based e-commerce recommendation system that harnesses style and shopping cart information to enhance existing transformer-based sequential product recommendation systems. Style4Rec represents a significant step forward in personalized e-commerce recommendations, outperforming benchmarks across various evaluation metrics. Style4Rec resulted in notable improvements: HR@5 increased from 0.681 to 0.735, NDCG@5 increased from 0.594 to 0.674, and MRR@5 increased from 0.559 to 0.654. We tested our model using an e-commerce dataset from our partnering company and found that it exceeded established transformer-based sequential recommendation benchmarks across various evaluation metrics. Thus, Style4Rec presents a significant step forward in personalized e-commerce recommendation systems.
comment: 9 pages, 6 images, 4 tables
☆ Rational Tuning of LLM Cascades via Probabilistic Modeling
Understanding the reliability of large language models (LLMs) has recently garnered significant attention. Given LLMs' propensity to hallucinate, as well as their high sensitivity to prompt design, it is already challenging to predict the performance of an individual LLM. However, the problem becomes more complex for compound LLM systems such as cascades, where in addition to each model's standalone performance, we must understand how the error rates of different models interact. In this paper, we present a probabilistic model for the joint performance distribution of a sequence of LLMs, which enables a framework for rationally tuning the confidence thresholds of a LLM cascade using continuous optimization. Compared to selecting confidence thresholds using grid search, our parametric Markov-copula model significantly improves runtime scaling with respect to the length of the cascade and the desired resolution of the cost-error curve, turning them from intractable into low-order polynomial. In addition, the optimal thresholds computed using our continuous optimization-based algorithm increasingly outperform those found via grid search as cascade length grows, improving the area under the cost-error curve by 1.9% on average for cascades consisting of at least three models. Overall, our Markov-copula model provides a rational basis for tuning LLM cascade performance and points to the potential of probabilistic methods in analyzing LLM systems.
Prompt-CAM: A Simpler Interpretable Transformer for Fine-Grained Analysis
We present a simple usage of pre-trained Vision Transformers (ViTs) for fine-grained analysis, aiming to identify and localize the traits that distinguish visually similar categories, such as different bird species or dog breeds. Pre-trained ViTs such as DINO have shown remarkable capabilities to extract localized, informative features. However, using saliency maps like Grad-CAM can hardly point out the traits: they often locate the whole object by a blurred, coarse heatmap, not traits. We propose a novel approach Prompt Class Attention Map (Prompt-CAM) to the rescue. Prompt-CAM learns class-specific prompts to a pre-trained ViT and uses the corresponding outputs for classification. To classify an image correctly, the true-class prompt must attend to the unique image patches not seen in other classes' images, i.e., traits. As such, the true class's multi-head attention maps reveal traits and their locations. Implementation-wise, Prompt-CAM is almost a free lunch by simply modifying the prediction head of Visual Prompt Tuning (VPT). This makes Prompt-CAM fairly easy to train and apply, sharply contrasting other interpretable methods that design specific models and training processes. It is even simpler than the recently published INterpretable TRansformer (INTR), whose encoder-decoder architecture prevents it from leveraging pre-trained ViTs. Extensive empirical studies on a dozen datasets from various domains (e.g., birds, fishes, insects, fungi, flowers, food, and cars) validate Prompt-CAM superior interpretation capability.
☆ Neural Honeytrace: A Robust Plug-and-Play Watermarking Framework against Model Extraction Attacks
Developing high-performance deep learning models is resource-intensive, leading model owners to utilize Machine Learning as a Service (MLaaS) platforms instead of publicly releasing their models. However, malicious users may exploit query interfaces to execute model extraction attacks, reconstructing the target model's functionality locally. While prior research has investigated triggerable watermarking techniques for asserting ownership, existing methods face significant challenges: (1) most approaches require additional training, resulting in high overhead and limited flexibility, and (2) they often fail to account for advanced attackers, leaving them vulnerable to adaptive attacks. In this paper, we propose Neural Honeytrace, a robust plug-and-play watermarking framework against model extraction attacks. We first formulate a watermark transmission model from an information-theoretic perspective, providing an interpretable account of the principles and limitations of existing triggerable watermarking. Guided by the model, we further introduce: (1) a similarity-based training-free watermarking method for plug-and-play and flexible watermarking, and (2) a distribution-based multi-step watermark information transmission strategy for robust watermarking. Comprehensive experiments on four datasets demonstrate that Neural Honeytrace outperforms previous methods in efficiency and resisting adaptive attacks. Neural Honeytrace reduces the average number of samples required for a worst-case t-Test-based copyright claim from $12,000$ to $200$ with zero training cost.
☆ On Learning Informative Trajectory Embeddings for Imitation, Classification and Regression AAMAS 2025
In real-world sequential decision making tasks like autonomous driving, robotics, and healthcare, learning from observed state-action trajectories is critical for tasks like imitation, classification, and clustering. For example, self-driving cars must replicate human driving behaviors, while robots and healthcare systems benefit from modeling decision sequences, whether or not they come from expert data. Existing trajectory encoding methods often focus on specific tasks or rely on reward signals, limiting their ability to generalize across domains and tasks. Inspired by the success of embedding models like CLIP and BERT in static domains, we propose a novel method for embedding state-action trajectories into a latent space that captures the skills and competencies in the dynamic underlying decision-making processes. This method operates without the need for reward labels, enabling better generalization across diverse domains and tasks. Our contributions are threefold: (1) We introduce a trajectory embedding approach that captures multiple abilities from state-action data. (2) The learned embeddings exhibit strong representational power across downstream tasks, including imitation, classification, clustering, and regression. (3) The embeddings demonstrate unique properties, such as controlling agent behaviors in IQ-Learn and an additive structure in the latent space. Experimental results confirm that our method outperforms traditional approaches, offering more flexible and powerful trajectory representations for various applications. Our code is available at https://github.com/Erasmo1015/vte.
comment: AAMAS 2025
☆ SOP-Agent: Empower General Purpose AI Agent with Domain-Specific SOPs
Despite significant advancements in general-purpose AI agents, several challenges still hinder their practical application in real-world scenarios. First, the limited planning capabilities of Large Language Models (LLM) restrict AI agents from effectively solving complex tasks that require long-horizon planning. Second, general-purpose AI agents struggle to efficiently utilize domain-specific knowledge and human expertise. In this paper, we introduce the Standard Operational Procedure-guided Agent (SOP-agent), a novel framework for constructing domain-specific agents through pseudocode-style Standard Operational Procedures (SOPs) written in natural language. Formally, we represent a SOP as a decision graph, which is traversed to guide the agent in completing tasks specified by the SOP. We conduct extensive experiments across tasks in multiple domains, including decision-making, search and reasoning, code generation, data cleaning, and grounded customer service. The SOP-agent demonstrates excellent versatility, achieving performance superior to general-purpose agent frameworks and comparable to domain-specific agent systems. Additionally, we introduce the Grounded Customer Service Benchmark, the first benchmark designed to evaluate the grounded decision-making capabilities of AI agents in customer service scenarios based on SOPs.
comment: 35 pages, 5 figures
☆ Shape-Based Single Object Classification Using Ensemble Method Classifiers
Nowadays, more and more images are available. Annotation and retrieval of the images pose classification problems, where each class is defined as the group of database images labelled with a common semantic label. Various systems have been proposed for content-based retrieval, as well as for image classification and indexing. In this paper, a hierarchical classification framework has been proposed for bridging the semantic gap effectively and achieving multi-category image classification. A well known pre-processing and post-processing method was used and applied to three problems; image segmentation, object identification and image classification. The method was applied to classify single object images from Amazon and Google datasets. The classification was tested for four different classifiers; BayesNetwork (BN), Random Forest (RF), Bagging and Vote. The estimated classification accuracies ranged from 20% to 99% (using 10-fold cross validation). The Bagging classifier presents the best performance, followed by the Random Forest classifier.
☆ A Study of In-Context-Learning-Based Text-to-SQL Errors
Large language models (LLMs) have been adopted to perform text-to-SQL tasks, utilizing their in-context learning (ICL) capability to translate natural language questions into structured query language (SQL). However, such a technique faces correctness problems and requires efficient repairing solutions. In this paper, we conduct the first comprehensive study of text-to-SQL errors. Our study covers four representative ICL-based techniques, five basic repairing methods, two benchmarks, and two LLM settings. We find that text-to-SQL errors are widespread and summarize 29 error types of 7 categories. We also find that existing repairing attempts have limited correctness improvement at the cost of high computational overhead with many mis-repairs. Based on the findings, we propose MapleRepair, a novel text-to-SQL error detection and repairing framework. The evaluation demonstrates that MapleRepair outperforms existing solutions by repairing 13.8% more queries with neglectable mis-repairs and 67.4% less overhead.
☆ Understanding Mental Health Content on Social Media and Its Effect Towards Suicidal Ideation
This review underscores the critical need for effective strategies to identify and support individuals with suicidal ideation, exploiting technological innovations in ML and DL to further suicide prevention efforts. The study details the application of these technologies in analyzing vast amounts of unstructured social media data to detect linguistic patterns, keywords, phrases, tones, and contextual cues associated with suicidal thoughts. It explores various ML and DL models like SVMs, CNNs, LSTM, neural networks, and their effectiveness in interpreting complex data patterns and emotional nuances within text data. The review discusses the potential of these technologies to serve as a life-saving tool by identifying at-risk individuals through their digital traces. Furthermore, it evaluates the real-world effectiveness, limitations, and ethical considerations of employing these technologies for suicide prevention, stressing the importance of responsible development and usage. The study aims to fill critical knowledge gaps by analyzing recent studies, methodologies, tools, and techniques in this field. It highlights the importance of synthesizing current literature to inform practical tools and suicide prevention efforts, guiding innovation in reliable, ethical systems for early intervention. This research synthesis evaluates the intersection of technology and mental health, advocating for the ethical and responsible application of ML, DL, and NLP to offer life-saving potential worldwide while addressing challenges like generalizability, biases, privacy, and the need for further research to ensure these technologies do not exacerbate existing inequities and harms.
☆ To Retrieve or Not to Retrieve? Uncertainty Detection for Dynamic Retrieval Augmented Generation
Retrieval-Augmented Generation equips large language models with the capability to retrieve external knowledge, thereby mitigating hallucinations by incorporating information beyond the model's intrinsic abilities. However, most prior works have focused on invoking retrieval deterministically, which makes it unsuitable for tasks such as long-form question answering. Instead, dynamically performing retrieval by invoking it only when the underlying LLM lacks the required knowledge can be more efficient. In this context, we delve deeper into the question, "To Retrieve or Not to Retrieve?" by exploring multiple uncertainty detection methods. We evaluate these methods for the task of long-form question answering, employing dynamic retrieval, and present our comparisons. Our findings suggest that uncertainty detection metrics, such as Degree Matrix Jaccard and Eccentricity, can reduce the number of retrieval calls by almost half, with only a slight reduction in question-answering accuracy.
☆ LAVCap: LLM-based Audio-Visual Captioning using Optimal Transport ICASSP 2025
Automated audio captioning is a task that generates textual descriptions for audio content, and recent studies have explored using visual information to enhance captioning quality. However, current methods often fail to effectively fuse audio and visual data, missing important semantic cues from each modality. To address this, we introduce LAVCap, a large language model (LLM)-based audio-visual captioning framework that effectively integrates visual information with audio to improve audio captioning performance. LAVCap employs an optimal transport-based alignment loss to bridge the modality gap between audio and visual features, enabling more effective semantic extraction. Additionally, we propose an optimal transport attention module that enhances audio-visual fusion using an optimal transport assignment map. Combined with the optimal training strategy, experimental results demonstrate that each component of our framework is effective. LAVCap outperforms existing state-of-the-art methods on the AudioCaps dataset, without relying on large datasets or post-processing. Code is available at https://github.com/NAVER-INTEL-Co-Lab/gaudi-lavcap.
comment: 5 pages, 2 figures; Accepted to ICASSP 2025
☆ SEAL: Entangled White-box Watermarks on Low-Rank Adaptation
Recently, LoRA and its variants have become the de facto strategy for training and sharing task-specific versions of large pretrained models, thanks to their efficiency and simplicity. However, the issue of copyright protection for LoRA weights, especially through watermark-based techniques, remains underexplored. To address this gap, we propose SEAL (SEcure wAtermarking on LoRA weights), the universal whitebox watermarking for LoRA. SEAL embeds a secret, non-trainable matrix between trainable LoRA weights, serving as a passport to claim ownership. SEAL then entangles the passport with the LoRA weights through training, without extra loss for entanglement, and distributes the finetuned weights after hiding the passport. When applying SEAL, we observed no performance degradation across commonsense reasoning, textual/visual instruction tuning, and text-to-image synthesis tasks. We demonstrate that SEAL is robust against a variety of known attacks: removal, obfuscation, and ambiguity attacks.
comment: 26 pages, 16 tables, 9 figures, initial version
☆ Text Semantics to Flexible Design: A Residential Layout Generation Method Based on Stable Diffusion Model
Flexibility in the AI-based residential layout design remains a significant challenge, as traditional methods like rule-based heuristics and graph-based generation often lack flexibility and require substantial design knowledge from users. To address these limitations, we propose a cross-modal design approach based on the Stable Diffusion model for generating flexible residential layouts. The method offers multiple input types for learning objectives, allowing users to specify both boundaries and layouts. It incorporates natural language as design constraints and introduces ControlNet to enable stable layout generation through two distinct pathways. We also present a scheme that encapsulates design expertise within a knowledge graph and translates it into natural language, providing an interpretable representation of design knowledge. This comprehensibility and diversity of input options enable professionals and non-professionals to directly express design requirements, enhancing flexibility and controllability. Finally, experiments verify the flexibility of the proposed methods under multimodal constraints better than state-of-the-art models, even when specific semantic information about room areas or connections is incomplete.
☆ Large Language Model is Secretly a Protein Sequence Optimizer
We consider the protein sequence engineering problem, which aims to find protein sequences with high fitness levels, starting from a given wild-type sequence. Directed evolution has been a dominating paradigm in this field which has an iterative process to generate variants and select via experimental feedback. We demonstrate large language models (LLMs), despite being trained on massive texts, are secretly protein sequence optimizers. With a directed evolutionary method, LLM can perform protein engineering through Pareto and experiment-budget constrained optimization, demonstrating success on both synthetic and experimental fitness landscapes.
comment: Preprint
☆ Perspective Transition of Large Language Models for Solving Subjective Tasks
Large language models (LLMs) have revolutionized the field of natural language processing, enabling remarkable progress in various tasks. Different from objective tasks such as commonsense reasoning and arithmetic question-answering, the performance of LLMs on subjective tasks is still limited, where the perspective on the specific problem plays crucial roles for better interpreting the context and giving proper response. For example, in certain scenarios, LLMs may perform better when answering from an expert role perspective, potentially eliciting their relevant domain knowledge. In contrast, in some scenarios, LLMs may provide more accurate responses when answering from a third-person standpoint, enabling a more comprehensive understanding of the problem and potentially mitigating inherent biases. In this paper, we propose Reasoning through Perspective Transition (RPT), a method based on in-context learning that enables LLMs to dynamically select among direct, role, and third-person perspectives for the best way to solve corresponding subjective problem. Through extensive experiments on totally 12 subjective tasks by using both closed-source and open-source LLMs including GPT-4, GPT-3.5, Llama-3, and Qwen-2, our method outperforms widely used single fixed perspective based methods such as chain-of-thought prompting and expert prompting, highlights the intricate ways that LLMs can adapt their perspectives to provide nuanced and contextually appropriate responses for different problems.
☆ Clone-Robust AI Alignment
A key challenge in training Large Language Models (LLMs) is properly aligning them with human preferences. Reinforcement Learning with Human Feedback (RLHF) uses pairwise comparisons from human annotators to train reward functions and has emerged as a popular alignment method. However, input datasets in RLHF are not necessarily balanced in the types of questions and answers that are included. Therefore, we want RLHF algorithms to perform well even when the set of alternatives is not uniformly distributed. Drawing on insights from social choice theory, we introduce robustness to approximate clones, a desirable property of RLHF algorithms which requires that adding near-duplicate alternatives does not significantly change the learned reward function. We first demonstrate that the standard RLHF algorithm based on regularized maximum likelihood estimation (MLE) fails to satisfy this property. We then propose the weighted MLE, a new RLHF algorithm that modifies the standard regularized MLE by weighting alternatives based on their similarity to other alternatives. This new algorithm guarantees robustness to approximate clones while preserving desirable theoretical properties.
☆ AI-based Identity Fraud Detection: A Systematic Review
With the rapid development of digital services, a large volume of personally identifiable information (PII) is stored online and is subject to cyberattacks such as Identity fraud. Most recently, the use of Artificial Intelligence (AI) enabled deep fake technologies has significantly increased the complexity of identity fraud. Fraudsters may use these technologies to create highly sophisticated counterfeit personal identification documents, photos and videos. These advancements in the identity fraud landscape pose challenges for identity fraud detection and society at large. There is a pressing need to review and understand identity fraud detection methods, their limitations and potential solutions. This research aims to address this important need by using the well-known systematic literature review method. This paper reviewed a selected set of 43 papers across 4 major academic literature databases. In particular, the review results highlight the two types of identity fraud prevention and detection methods, in-depth and open challenges. The results were also consolidated into a taxonomy of AI-based identity fraud detection and prevention methods including key insights and trends. Overall, this paper provides a foundational knowledge base to researchers and practitioners for further research and development in this important area of digital identity fraud.
☆ Foundations of Large Language Models
This is a book about large language models. As indicated by the title, it primarily focuses on foundational concepts rather than comprehensive coverage of all cutting-edge technologies. The book is structured into four main chapters, each exploring a key area: pre-training, generative models, prompting techniques, and alignment methods. It is intended for college students, professionals, and practitioners in natural language processing and related fields, and can serve as a reference for anyone interested in large language models.
☆ Interpretable Droplet Digital PCR Assay for Trustworthy Molecular Diagnostics
Accurate molecular quantification is essential for advancing research and diagnostics in fields such as infectious diseases, cancer biology, and genetic disorders. Droplet digital PCR (ddPCR) has emerged as a gold standard for achieving absolute quantification. While computational ddPCR technologies have advanced significantly, achieving automatic interpretation and consistent adaptability across diverse operational environments remains a challenge. To address these limitations, we introduce the intelligent interpretable droplet digital PCR (I2ddPCR) assay, a comprehensive framework integrating front-end predictive models (for droplet segmentation and classification) with GPT-4o multimodal large language model (MLLM, for context-aware explanations and recommendations) to automate and enhance ddPCR image analysis. This approach surpasses the state-of-the-art models, affording 99.05% accuracy in processing complex ddPCR images containing over 300 droplets per image with varying signal-to-noise ratios (SNRs). By combining specialized neural networks and large language models, the I2ddPCR assay offers a robust and adaptable solution for absolute molecular quantification, achieving a sensitivity capable of detecting low-abundance targets as low as 90.32 copies/{\mu}L. Furthermore, it improves model's transparency through detailed explanation and troubleshooting guidance, empowering users to make informed decisions. This innovative framework has the potential to benefit molecular diagnostics, disease research, and clinical applications, especially in resource-constrained settings.
☆ Adaptive Law-Based Transformation (ALT): A Lightweight Feature Representation for Time Series Classification
Time series classification (TSC) is fundamental in numerous domains, including finance, healthcare, and environmental monitoring. However, traditional TSC methods often struggle with the inherent complexity and variability of time series data. Building on our previous work with the linear law-based transformation (LLT) - which improved classification accuracy by transforming the feature space based on key data patterns - we introduce adaptive law-based transformation (ALT). ALT enhances LLT by incorporating variable-length shifted time windows, enabling it to capture distinguishing patterns of various lengths and thereby handle complex time series more effectively. By mapping features into a linearly separable space, ALT provides a fast, robust, and transparent solution that achieves state-of-the-art performance with only a few hyperparameters.
comment: 8 pages, 1 figure, 5 tables
♻ ☆ Meaning-Typed Programming: Language-level Abstractions and Runtime for GenAI Applications
Software is rapidly evolving from being programmed with traditional logical code, to neuro-integrated applications that leverage generative AI and large language models (LLMs) for application functionality. This shift increases the complexity of building applications, as developers now must reasoning about, program, and prompt LLMs. Despite efforts to create tools to assist with prompt engineering, these solutions often introduce additional layers of complexity to the development of neuro-integrated applications. This paper proposes meaning-typed programming (MTP), a novel approach to simplify the creation of neuro-integrated applications by introducing new language-level abstractions that hide the complexities of LLM integration. Our key insight is that typical conventional code already possesses a high level of semantic richness that can be automatically reasoned about, as it is designed to be readable and maintainable by humans. Leveraging this insight, we conceptualize LLMs as meaning-typed code constructs and introduce a by abstraction at the language level, MT-IR, a new meaning-based intermediate representation at the compiler level, and MT Runtime, an automated run-time engine for LLM integration and operations. We implement MTP in a production-grade Python super-set language called Jac and perform an extensive evaluation. Our results demonstrate that MTP not only simplifies the development process but also meets or exceeds the efficacy of state-of-the-art manual and tool-assisted prompt engineering techniques in terms of accuracy and usability.
♻ ☆ Using Machine Learning to Discover Parsimonious and Physically-Interpretable Representations of Catchment-Scale Rainfall-Runoff Dynamics
Despite the excellent real-world predictive performance of modern machine learning (ML) methods, many scientists remain hesitant to discard traditional physical-conceptual (PC) approaches due mainly to their relative interpretability, which contributes to credibility during decision-making. In this context, a currently underexplored aspect of ML is how to develop minimally-optimal representations that can facilitate better insight regarding system functioning. Regardless of how this is achieved, it is arguably true that parsimonious representations better support the advancement of scientific understanding. Our own view is that ML-based modeling of geoscientific systems should be based in the use of computational units that are fundamentally interpretable by design. This paper continues our exploration of how the strengths of ML can be exploited in the service of better understanding via scientific investigation. Here, we use the Mass Conserving Perceptron (MCP) as the fundamental computational unit in a generic network architecture consisting of nodes arranged in series and parallel to explore several generic and important issues related to the use of observational data for constructing input-state-output models of dynamical systems. In the context of lumped catchment modeling, we show that physical interpretability and excellent predictive performance can both be achieved using a relatively parsimonious distributed-state multiple-flow-path network with context-dependent gating and information sharing across the nodes, suggesting that MCP-based modeling can play a significant role in application of ML to geoscientific investigation.
comment: 74 Pages, 4 Tables, 13 Figures, 11 Tables and 11 Figures in Supplementary Materials
♻ ☆ NL2KQL: From Natural Language to Kusto Query
Data is growing rapidly in volume and complexity. Proficiency in database query languages is pivotal for crafting effective queries. As coding assistants become more prevalent, there is significant opportunity to enhance database query languages. The Kusto Query Language (KQL) is a widely used query language for large semi-structured data such as logs, telemetries, and time-series for big data analytics platforms. This paper introduces NL2KQL an innovative framework that uses large language models (LLMs) to convert natural language queries (NLQs) to KQL queries. The proposed NL2KQL framework includes several key components: Schema Refiner which narrows down the schema to its most pertinent elements; the Few-shot Selector which dynamically selects relevant examples from a few-shot dataset; and the Query Refiner which repairs syntactic and semantic errors in KQL queries. Additionally, this study outlines a method for generating large datasets of synthetic NLQ-KQL pairs which are valid within a specific database contexts. To validate NL2KQL's performance, we utilize an array of online (based on query execution) and offline (based on query parsing) metrics. Through ablation studies, the significance of each framework component is examined, and the datasets used for benchmarking are made publicly available. This work is the first of its kind and is compared with available baselines to demonstrate its effectiveness.
♻ ☆ Frechet Music Distance: A Metric For Generative Symbolic Music Evaluation
In this paper we introduce the Frechet Music Distance (FMD), a novel evaluation metric for generative symbolic music models, inspired by the Frechet Inception Distance (FID) in computer vision and Frechet Audio Distance (FAD) in generative audio. FMD calculates the distance between distributions of reference and generated symbolic music embeddings, capturing abstract musical features. We validate FMD across several datasets and models. Results indicate that FMD effectively differentiates model quality, providing a domain-specific metric for evaluating symbolic music generation, and establishing a reproducible standard for future research in symbolic music modeling.
♻ ☆ Dynamics of Moral Behavior in Heterogeneous Populations of Learning Agents AAAI
Growing concerns about safety and alignment of AI systems highlight the importance of embedding moral capabilities in artificial agents: a promising solution is the use of learning from experience, i.e., Reinforcement Learning. In multi-agent (social) environments, complex population-level phenomena may emerge from interactions between individual learning agents. Many of the existing studies rely on simulated social dilemma environments to study the interactions of independent learning agents; however, they tend to ignore the moral heterogeneity that is likely to be present in societies of agents in practice. For example, at different points in time a single learning agent may face opponents who are consequentialist (i.e., focused on maximizing outcomes over time), norm-based (i.e., conforming to specific norms), or virtue-based (i.e., considering a combination of different virtues). The extent to which agents' co-development may be impacted by such moral heterogeneity in populations is not well understood. In this paper, we present a study of the learning dynamics of morally heterogeneous populations interacting in a social dilemma setting. Using an Iterated Prisoner's Dilemma environment with a partner selection mechanism, we investigate the extent to which the prevalence of diverse moral agents in populations affects individual agents' learning behaviors and emergent population-level outcomes. We observe several types of non-trivial interactions between pro-social and anti-social agents, and find that certain types of moral agents are able to steer selfish agents towards more cooperative behavior.
comment: Presented at AIES 2024 (7th AAAI/ACM Conference on AI, Ethics, and Society - San Jose, CA, USA) - see https://ojs.aaai.org/index.php/AIES/article/view/31736
♻ ☆ Convex Markov Games: A Framework for Creativity, Imitation, Fairness, and Safety in Multiagent Learning
Behavioral diversity, expert imitation, fairness, safety goals and others give rise to preferences in sequential decision making domains that do not decompose additively across time. We introduce the class of convex Markov games that allow general convex preferences over occupancy measures. Despite infinite time horizon and strictly higher generality than Markov games, pure strategy Nash equilibria exist. Furthermore, equilibria can be approximated empirically by performing gradient descent on an upper bound of exploitability. Our experiments reveal novel solutions to classic repeated normal-form games, find fair solutions in a repeated asymmetric coordination game, and prioritize safe long-term behavior in a robot warehouse environment. In the prisoner's dilemma, our algorithm leverages transient imitation to find a policy profile that deviates from observed human play only slightly, yet achieves higher per-player utility while also being three orders of magnitude less exploitable.
♻ ☆ A Systems Thinking Approach to Algorithmic Fairness
Systems thinking provides us with a way to model the algorithmic fairness problem by allowing us to encode prior knowledge and assumptions about where we believe bias might exist in the data generating process. We can then encode these beliefs as a series of causal graphs, enabling us to link AI/ML systems to politics and the law. This allows us to combine techniques from machine learning, causal inference, and system dynamics in order to capture different emergent aspects of the fairness problem. We can use systems thinking to help policymakers on both sides of the political aisle to understand the complex trade-offs that exist from different types of fairness policies, providing a sociotechnical foundation for designing AI policy that is aligned to their political agendas.
comment: This paper has been submitted to the 2025 ACM FAccT conference for review
♻ ☆ A Comparative Study on Multi-task Uncertainty Quantification in Semantic Segmentation and Monocular Depth Estimation
Deep neural networks excel in perception tasks such as semantic segmentation and monocular depth estimation, making them indispensable in safety-critical applications like autonomous driving and industrial inspection. However, they often suffer from overconfidence and poor explainability, especially for out-of-domain data. While uncertainty quantification has emerged as a promising solution to these challenges, multi-task settings have yet to be explored. In an effort to shed light on this, we evaluate Monte Carlo Dropout, Deep Sub-Ensembles, and Deep Ensembles for joint semantic segmentation and monocular depth estimation. Thereby, we reveal that Deep Ensembles stand out as the preferred choice, particularly in out-of-domain scenarios, and show the potential benefit of multi-task learning with regard to the uncertainty quality in comparison to solving both tasks separately. Additionally, we highlight the impact of employing different uncertainty thresholds to classify pixels as certain or uncertain, with the median uncertainty emerging as a robust default.
comment: This manuscript is an extended version of a previously published conference paper and is currently in review for a journal
♻ ☆ A Comprehensive Survey of Foundation Models in Medicine
Foundation models (FMs) are large-scale deep learning models trained on massive datasets, often using self-supervised learning techniques. These models serve as a versatile base for a wide range of downstream tasks, including those in medicine and healthcare. FMs have demonstrated remarkable success across multiple healthcare domains. However, existing surveys in this field do not comprehensively cover all areas where FMs have made significant strides. In this survey, we present a comprehensive review of FMs in medicine, focusing on their evolution, learning strategies, flagship models, applications, and associated challenges. We examine how prominent FMs, such as the BERT and GPT families, are transforming various aspects of healthcare, including clinical large language models, medical image analysis, and omics research. Additionally, we provide a detailed taxonomy of FM-enabled healthcare applications, spanning clinical natural language processing, medical computer vision, graph learning, and other biology- and omics- related tasks. Despite the transformative potentials of FMs, they also pose unique challenges. This survey delves into these challenges and highlights open research questions and lessons learned to guide researchers and practitioners. Our goal is to provide valuable insights into the capabilities of FMs in health, facilitating responsible deployment and mitigating associated risks.
comment: Currently under review in IEEE REVIEWS IN BIOMEDICAL ENGINEERING
♻ ☆ Improving Zero-Shot Object-Level Change Detection by Incorporating Visual Correspondence
Detecting object-level changes between two images across possibly different views is a core task in many applications that involve visual inspection or camera surveillance. Existing change-detection approaches suffer from three major limitations: (1) lack of evaluation on image pairs that contain no changes, leading to unreported false positive rates; (2) lack of correspondences (i.e., localizing the regions before and after a change); and (3) poor zero-shot generalization across different domains. To address these issues, we introduce a novel method that leverages change correspondences (a) during training to improve change detection accuracy, and (b) at test time, to minimize false positives. That is, we harness the supervision labels of where an object is added or removed to supervise change detectors, improving their accuracy over previous work by a large margin. Our work is also the first to predict correspondences between pairs of detected changes using estimated homography and the Hungarian algorithm. Our model demonstrates superior performance over existing methods, achieving state-of-the-art results in change detection and change correspondence accuracy across both in-distribution and zero-shot benchmarks.
♻ ☆ Hybrid Approaches for Moral Value Alignment in AI Agents: a Manifesto
Increasing interest in ensuring the safety of next-generation Artificial Intelligence (AI) systems calls for novel approaches to embedding morality into autonomous agents. This goal differs qualitatively from traditional task-specific AI methodologies. In this paper, we provide a systematization of existing approaches to the problem of introducing morality in machines - modelled as a continuum. Our analysis suggests that popular techniques lie at the extremes of this continuum - either being fully hard-coded into top-down, explicit rules, or entirely learned in a bottom-up, implicit fashion with no direct statement of any moral principle (this includes learning from human feedback, as applied to the training and finetuning of large language models, or LLMs). Given the relative strengths and weaknesses of each type of methodology, we argue that more hybrid solutions are needed to create adaptable and robust, yet controllable and interpretable agentic systems. To that end, this paper discusses both the ethical foundations (including deontology, consequentialism and virtue ethics) and implementations of morally aligned AI systems. We present a series of case studies that rely on intrinsic rewards, moral constraints or textual instructions, applied to either pure-Reinforcement Learning or LLM-based agents. By analysing these diverse implementations under one framework, we compare their relative strengths and shortcomings in developing morally aligned AI systems. We then discuss strategies for evaluating the effectiveness of moral learning agents. Finally, we present open research questions and implications for the future of AI safety and ethics which are emerging from this hybrid framework.
♻ ☆ Monte Carlo Tree Search for Comprehensive Exploration in LLM-Based Automatic Heuristic Design
Handcrafting heuristics for solving complex planning tasks (e.g., NP-hard combinatorial optimization (CO) problems) is a common practice but requires extensive domain knowledge. Recently, Large Language Model (LLM)-based automatic heuristics design (AHD) methods have shown promise in generating high-quality heuristics without manual intervention. Existing LLM-based AHD methods employ a population to maintain a fixed number of top-performing LLM-generated heuristics and introduce evolutionary computation (EC) to enhance the population iteratively. However, the population-based procedure brings greedy properties, often resulting in convergence to local optima. Instead, to more comprehensively explore the space of heuristics, we propose using Monte Carlo Tree Search (MCTS) for LLM-based heuristic evolution while preserving all LLM-generated heuristics in a tree structure. With a novel thought-alignment process and an exploration-decay technique, the proposed MCTS-AHD method delivers significantly higher-quality heuristics on various complex tasks. Our code is available at https://github.com/zz1358m/MCTS-AHD-master.
♻ ☆ ReFactor GNNs: Revisiting Factorisation-based Models from a Message-Passing Perspective NeurIPS 2022
Factorisation-based Models (FMs), such as DistMult, have enjoyed enduring success for Knowledge Graph Completion (KGC) tasks, often outperforming Graph Neural Networks (GNNs). However, unlike GNNs, FMs struggle to incorporate node features and generalise to unseen nodes in inductive settings. Our work bridges the gap between FMs and GNNs by proposing ReFactor GNNs. This new architecture draws upon both modelling paradigms, which previously were largely thought of as disjoint. Concretely, using a message-passing formalism, we show how FMs can be cast as GNNs by reformulating the gradient descent procedure as message-passing operations, which forms the basis of our ReFactor GNNs. Across a multitude of well-established KGC benchmarks, our ReFactor GNNs achieve comparable transductive performance to FMs, and state-of-the-art inductive performance while using an order of magnitude fewer parameters.
comment: 36th Conference on Neural Information Processing Systems (NeurIPS 2022)
♻ ☆ Bayesian Low-Rank LeArning (Bella): A Practical Approach to Bayesian Neural Networks AAAI'2025
Computational complexity of Bayesian learning is impeding its adoption in practical, large-scale tasks. Despite demonstrations of significant merits such as improved robustness and resilience to unseen or out-of-distribution inputs over their non- Bayesian counterparts, their practical use has faded to near insignificance. In this study, we introduce an innovative framework to mitigate the computational burden of Bayesian neural networks (BNNs). Our approach follows the principle of Bayesian techniques based on deep ensembles, but significantly reduces their cost via multiple low-rank perturbations of parameters arising from a pre-trained neural network. Both vanilla version of ensembles as well as more sophisticated schemes such as Bayesian learning with Stein Variational Gradient Descent (SVGD), previously deemed impractical for large models, can be seamlessly implemented within the proposed framework, called Bayesian Low-Rank LeArning (Bella). In a nutshell, i) Bella achieves a dramatic reduction in the number of trainable parameters required to approximate a Bayesian posterior; and ii) it not only maintains, but in some instances, surpasses the performance of conventional Bayesian learning methods and non-Bayesian baselines. Our results with large-scale tasks such as ImageNet, CAMELYON17, DomainNet, VQA with CLIP, LLaVA demonstrate the effectiveness and versatility of Bella in building highly scalable and practical Bayesian deep models for real-world applications.
comment: This paper is accepted in AAAI'2025
♻ ☆ Modeling Time-Variant Responses of Optical Compressors with Selective State Space Models
This paper presents a method for modeling optical dynamic range compressors using deep neural networks with Selective State Space models. The proposed approach surpasses previous methods based on recurrent layers by employing a Selective State Space block to encode the input audio. It features a refined technique integrating Feature-wise Linear Modulation and Gated Linear Units to adjust the network dynamically, conditioning the compression's attack and release phases according to external parameters. The proposed architecture is well-suited for low-latency and real-time applications, crucial in live audio processing. The method has been validated on the analog optical compressors TubeTech CL 1B and Teletronix LA-2A, which possess distinct characteristics. Evaluation is performed using quantitative metrics and subjective listening tests, comparing the proposed method with other state-of-the-art models. Results show that our black-box modeling methods outperform all others, achieving accurate emulation of the compression process for both seen and unseen settings during training. We further show a correlation between this accuracy and the sampling density of the control parameters in the dataset and identify settings with fast attack and slow release as the most challenging to emulate.
comment: Journal of the Audio Engineering Society
♻ ☆ Enhancing Few-Shot Image Classification through Learnable Multi-Scale Embedding and Attention Mechanisms
In the context of few-shot classification, the goal is to train a classifier using a limited number of samples while maintaining satisfactory performance. However, traditional metric-based methods exhibit certain limitations in achieving this objective. These methods typically rely on a single distance value between the query feature and support feature, thereby overlooking the contribution of shallow features. To overcome this challenge, we propose a novel approach in this paper. Our approach involves utilizing a multi-output embedding network that maps samples into distinct feature spaces. The proposed method extracts feature vectors at different stages, enabling the model to capture both global and abstract features. By utilizing these diverse feature spaces, our model enhances its performance. Moreover, employing a self-attention mechanism improves the refinement of features at each stage, leading to even more robust representations and improved overall performance. Furthermore, assigning learnable weights to each stage significantly improved performance and results. We conducted comprehensive evaluations on the MiniImageNet and FC100 datasets, specifically in the 5-way 1-shot and 5-way 5-shot scenarios. Additionally, we performed cross-domain tasks across eight benchmark datasets, achieving high accuracy in the testing domains. These evaluations demonstrate the efficacy of our proposed method in comparison to state-of-the-art approaches. https://github.com/FatemehAskari/MSENet
♻ ☆ Sines, Transient, Noise Neural Modeling of Piano Notes
This paper introduces a novel method for emulating piano sounds. We propose to exploit the sines, transient, and noise decomposition to design a differentiable spectral modeling synthesizer replicating piano notes. Three sub-modules learn these components from piano recordings and generate the corresponding harmonic, transient, and noise signals. Splitting the emulation into three independently trainable models reduces the modeling tasks' complexity. The quasi-harmonic content is produced using a differentiable sinusoidal model guided by physics-derived formulas, whose parameters are automatically estimated from audio recordings. The noise sub-module uses a learnable time-varying filter, and the transients are generated using a deep convolutional network. From singular notes, we emulate the coupling between different keys in trichords with a convolutional-based network. Results show the model matches the partial distribution of the target while predicting the energy in the higher part of the spectrum presents more challenges. The energy distribution in the spectra of the transient and noise components is accurate overall. While the model is more computationally and memory efficient, perceptual tests reveal limitations in accurately modeling the attack phase of notes. Despite this, it generally achieves perceptual accuracy in emulating single notes and trichords.
♻ ☆ Safe Control and Learning Using the Generalized Action Governor
This article introduces a general framework for safe control and learning based on the generalized action governor (AG). The AG is a supervisory scheme for augmenting a nominal closed-loop system with the ability of strictly handling prescribed safety constraints. In the first part of this article, we present a generalized AG methodology and analyze its key properties in a general setting. Then, we introduce tailored AG design approaches derived from the generalized methodology for linear and discrete systems. Afterward, we discuss the application of the generalized AG to facilitate safe online learning, which aims at safely evolving control parameters using real-time data to enhance control performance in uncertain systems. We present two safe learning algorithms based on, respectively, reinforcement learning and data-driven Koopman operator-based control integrated with the generalized AG to exemplify this application. Finally, we illustrate the developments with a numerical example.
comment: 22 pages, 4 figures, submitted to the International Journal of Control
♻ ☆ Silent Abandonment in Text-Based Contact Centers: Identifying, Quantifying, and Mitigating its Operational Impacts
In the quest to improve services, companies offer customers the option to interact with agents via texting. Such contact centers face unique challenges compared to traditional call centers, as measuring customer experience proxies like abandonment and patience involves uncertainty. A key source of this uncertainty is silent abandonment, where customers leave without notifying the system, wasting agent time and leaving their status unclear. Silent abandonment also obscures whether a customer was served or left. Our goals are to measure the magnitude of silent abandonment and mitigate its effects. Classification models show that 3%-70% of customers across 17 companies abandon silently. In one study, 71.3% of abandoning customers did so silently, reducing agent efficiency by 3.2% and system capacity by 15.3%, incurring $5,457 in annual costs per agent. We develop an expectation-maximization (EM) algorithm to estimate customer patience under uncertainty and identify influencing covariates. We find that companies should use classification models to estimate abandonment scope and our EM algorithm to assess patience. We suggest strategies to operationally mitigate the impact of silent abandonment by predicting suspected silent-abandonment behavior or changing service design. Specifically, we show that while allowing customers to write while waiting in the queue creates a missing data challenge, it also significantly increases patience and reduces service time, leading to reduced abandonment and lower staffing requirements.
comment: 75% of the paper is an updated version of arXiv:2304.11754
♻ ☆ A Multi-Modal Approach for Face Anti-Spoofing in Non-Calibrated Systems using Disparity Maps
Face recognition technologies are increasingly used in various applications, yet they are vulnerable to face spoofing attacks. These spoofing attacks often involve unique 3D structures, such as printed papers or mobile device screens. Although stereo-depth cameras can detect such attacks effectively, their high-cost limits their widespread adoption. Conversely, two-sensor systems without extrinsic calibration offer a cost-effective alternative but are unable to calculate depth using stereo techniques. In this work, we propose a method to overcome this challenge by leveraging facial attributes to derive disparity information and estimate relative depth for anti-spoofing purposes, using non-calibrated systems. We introduce a multi-modal anti-spoofing model, coined Disparity Model, that incorporates created disparity maps as a third modality alongside the two original sensor modalities. We demonstrate the effectiveness of the Disparity Model in countering various spoof attacks using a comprehensive dataset collected from the Intel RealSense ID Solution F455. Our method outperformed existing methods in the literature, achieving an Equal Error Rate (EER) of 1.71% and a False Negative Rate (FNR) of 2.77% at a False Positive Rate (FPR) of 1%. These errors are lower by 2.45% and 7.94% than the errors of the best comparison method, respectively. Additionally, we introduce a model ensemble that addresses 3D spoof attacks as well, achieving an EER of 2.04% and an FNR of 3.83% at an FPR of 1%. Overall, our work provides a state-of-the-art solution for the challenging task of anti-spoofing in non-calibrated systems that lack depth information.
♻ ☆ aiXcoder-7B: A Lightweight and Effective Large Language Model for Code Processing ICSE 2025
Large Language Models (LLMs) have been widely used in code completion, and researchers are focusing on scaling up LLMs to improve their accuracy. However, larger LLMs have lower inference efficiency, affecting developers' experience and productivity. In this paper, we propose a lightweight and effective LLM for code completion named aiXcoder-7B. Compared to existing LLMs, aiXcoder-7B achieves higher code completion accuracy while having smaller scales (i.e., 7 billion parameters). We attribute the superiority of aiXcoder-7B to three key factors: (1) Multi-objective training. We employ three training objectives, one of which is our proposed Structured Fill-In-the-Middle (SFIM). SFIM considers the syntax structures in code and effectively improves the performance of LLMs for code. (2) Diverse data sampling strategies. They consider inter-file relationships and enhance the capability of LLMs in understanding cross-file contexts. (3) Extensive high-quality data. We establish a rigorous data collection pipeline and consume a total of 1.2 trillion unique tokens for training aiXcoder-7B. This vast volume of data enables aiXcoder-7B to learn a broad distribution of code. We evaluate aiXcoder-7B in five popular code completion benchmarks and a new benchmark collected by this paper. The results show that aiXcoder-7B outperforms the latest six LLMs with similar sizes and even surpasses four larger LLMs (e.g., StarCoder2-15B and CodeLlama-34B), positioning aiXcoder-7B as a lightweight and effective LLM for academia and industry. Finally, we summarize three valuable insights for helping practitioners train the next generations of LLMs for code. aiXcoder-7B has been open-souced and gained significant attention. Until January 2025, aiXcoder-7B has received 2,226 GitHub Stars.
comment: (1) Accepted by the 47th International Conference on Software Engineering (ICSE 2025). (2) aiXcoder-7B is available at https://github.com/aixcoder-plugin/aiXcoder-7B
♻ ☆ AudioBERT: Audio Knowledge Augmented Language Model ICASSP 2025
Recent studies have identified that language models, pretrained on text-only datasets, often lack elementary visual knowledge, \textit{e.g.,} colors of everyday objects. Motivated by this observation, we ask whether a similar shortcoming exists in terms of the \textit{auditory} knowledge. To answer this question, we construct a new dataset called AuditoryBench, which consists of two novel tasks for evaluating auditory knowledge. Based on our analysis using the benchmark, we find that language models also suffer from a severe lack of auditory knowledge. To address this limitation, we propose AudioBERT, a novel method to augment the auditory knowledge of BERT through a retrieval-based approach. First, we detect auditory knowledge spans in prompts to query our retrieval model efficiently. Then, we inject audio knowledge into BERT and switch on low-rank adaptation for effective adaptation when audio knowledge is required. Our experiments demonstrate that AudioBERT is quite effective, achieving superior performance on the AuditoryBench. The dataset and code are available at \bulurl{https://github.com/HJ-Ok/AudioBERT}.
comment: 5 pages, 3 figures, ICASSP 2025
♻ ☆ Evaluating alignment between humans and neural network representations in image-based learning tasks
Humans represent scenes and objects in rich feature spaces, carrying information that allows us to generalise about category memberships and abstract functions with few examples. What determines whether a neural network model generalises like a human? We tested how well the representations of $86$ pretrained neural network models mapped to human learning trajectories across two tasks where humans had to learn continuous relationships and categories of natural images. In these tasks, both human participants and neural networks successfully identified the relevant stimulus features within a few trials, demonstrating effective generalisation. We found that while training dataset size was a core determinant of alignment with human choices, contrastive training with multi-modal data (text and imagery) was a common feature of currently publicly available models that predicted human generalisation. Intrinsic dimensionality of representations had different effects on alignment for different model types. Lastly, we tested three sets of human-aligned representations and found no consistent improvements in predictive accuracy compared to the baselines. In conclusion, pretrained neural networks can serve to extract representations for cognitive models, as they appear to capture some fundamental aspects of cognition that are transferable across tasks. Both our paradigms and modelling approach offer a novel way to quantify alignment between neural networks and humans and extend cognitive science into more naturalistic domains.
♻ ☆ Learning Constraint Network from Demonstrations via Positive-Unlabeled Learning with Memory Replay
Planning for a wide range of real-world tasks necessitates to know and write all constraints. However, instances exist where these constraints are either unknown or challenging to specify accurately. A possible solution is to infer the unknown constraints from expert demonstration. The majority of prior works limit themselves to learning simple linear constraints, or require strong knowledge of the true constraint parameterization or environmental model. To mitigate these problems, this paper presents a positive-unlabeled (PU) learning approach to infer a continuous, arbitrary and possibly nonlinear, constraint from demonstration. From a PU learning view, We treat all data in demonstrations as positive (feasible) data, and learn a (sub)-optimal policy to generate high-reward-winning but potentially infeasible trajectories, which serve as unlabeled data containing both feasible and infeasible states. Under an assumption on data distribution, a feasible-infeasible classifier (i.e., constraint model) is learned from the two datasets through a postprocessing PU learning technique. The entire method employs an iterative framework alternating between updating the policy, which generates and selects higher-reward policies, and updating the constraint model. Additionally, a memory buffer is introduced to record and reuse samples from previous iterations to prevent forgetting. The effectiveness of the proposed method is validated in two Mujoco environments, successfully inferring continuous nonlinear constraints and outperforming a baseline method in terms of constraint accuracy and policy safety.
♻ ☆ Focus On This, Not That! Steering LLMs With Adaptive Feature Specification
Despite the success of Instruction Tuning (IT) in training large language models (LLMs) to perform arbitrary user-specified tasks, these models often still leverage spurious or biased features learned from their training data, leading to undesired behaviours when deploying them in new contexts. In this work, we introduce Focus Instruction Tuning (FIT), which trains LLMs to condition their responses by focusing on specific features whilst ignoring others, leading to different behaviours based on what features are specified. Across several experimental settings, we show that focus-tuned models can be adaptively steered by focusing on different features at inference-time: for instance, robustness can be improved by focusing on task-causal features and ignoring spurious features, and social bias can be mitigated by ignoring demographic categories. Furthermore, FIT can steer behaviour in new contexts, generalising under distribution shift and to new unseen features at inference time, and thereby facilitating more robust, fair, and controllable LLM applications in real-world environments.
comment: 28pages, 14 figures
♻ ☆ Diffusion Models in Vision: A Survey
Denoising diffusion models represent a recent emerging topic in computer vision, demonstrating remarkable results in the area of generative modeling. A diffusion model is a deep generative model that is based on two stages, a forward diffusion stage and a reverse diffusion stage. In the forward diffusion stage, the input data is gradually perturbed over several steps by adding Gaussian noise. In the reverse stage, a model is tasked at recovering the original input data by learning to gradually reverse the diffusion process, step by step. Diffusion models are widely appreciated for the quality and diversity of the generated samples, despite their known computational burdens, i.e. low speeds due to the high number of steps involved during sampling. In this survey, we provide a comprehensive review of articles on denoising diffusion models applied in vision, comprising both theoretical and practical contributions in the field. First, we identify and present three generic diffusion modeling frameworks, which are based on denoising diffusion probabilistic models, noise conditioned score networks, and stochastic differential equations. We further discuss the relations between diffusion models and other deep generative models, including variational auto-encoders, generative adversarial networks, energy-based models, autoregressive models and normalizing flows. Then, we introduce a multi-perspective categorization of diffusion models applied in computer vision. Finally, we illustrate the current limitations of diffusion models and envision some interesting directions for future research.
comment: Accepted in IEEE Transactions on Pattern Analysis and Machine Intelligence. 25 pages, 3 figures
♻ ☆ Positive-Unlabeled Constraint Learning for Inferring Nonlinear Continuous Constraints Functions from Expert Demonstrations
Planning for diverse real-world robotic tasks necessitates to know and write all constraints. However, instances exist where these constraints are either unknown or challenging to specify accurately. A possible solution is to infer the unknown constraints from expert demonstration. This paper presents a novel two-step Positive-Unlabeled Constraint Learning (PUCL) algorithm to infer a continuous constraint function from demonstrations, without requiring prior knowledge of the true constraint parameterization or environmental model as existing works. We treat all data in demonstrations as positive (feasible) data, and learn a control policy to generate potentially infeasible trajectories, which serve as unlabeled data. The proposed two-step learning framework first identifies reliable infeasible data using a distance metric, and secondly learns a binary feasibility classifier (i.e., constraint function) from the feasible demonstrations and reliable infeasible data. The proposed method is flexible to learn complex-shaped constraint boundary and will not mistakenly classify demonstrations as infeasible as previous methods. The effectiveness of the proposed method is verified in four constrained environments, using a networked policy or a dynamical system policy. It successfully infers the continuous nonlinear constraints and outperforms other baseline methods in terms of constraint accuracy and policy safety. This work has been published in IEEE Robotics and Automation Letters (RA-L). Please refer to the final version at https://doi.org/10.1109/LRA.2024.3522756
♻ ☆ RAGBench: Explainable Benchmark for Retrieval-Augmented Generation Systems
Retrieval-Augmented Generation (RAG) has become a standard architectural pattern for incorporating domain-specific knowledge into user-facing chat applications powered by Large Language Models (LLMs). RAG systems are characterized by (1) a document retriever that queries a domain-specific corpus for context information relevant to an input query, and (2) an LLM that generates a response based on the provided query and context. However, comprehensive evaluation of RAG systems remains a challenge due to the lack of unified evaluation criteria and annotated datasets. In response, we introduce RAGBench: the first comprehensive, large-scale RAG benchmark dataset of 100k examples. It covers five unique industry-specific domains and various RAG task types. RAGBench examples are sourced from industry corpora such as user manuals, making it particularly relevant for industry applications. Further, we formalize the TRACe evaluation framework: a set of explainable and actionable RAG evaluation metrics applicable across all RAG domains. We release the labeled dataset at https://huggingface.co/datasets/rungalileo/ragbench. RAGBench explainable labels facilitate holistic evaluation of RAG systems, enabling actionable feedback for continuous improvement of production applications. Thorough extensive benchmarking, we find that LLM-based RAG evaluation methods struggle to compete with a finetuned RoBERTa model on the RAG evaluation task. We identify areas where existing approaches fall short and propose the adoption of RAGBench with TRACe towards advancing the state of RAG evaluation systems.
♻ ☆ Knowledge Retrieval Based on Generative AI
This study develops a question-answering system based on Retrieval-Augmented Generation (RAG) using Chinese Wikipedia and Lawbank as retrieval sources. Using TTQA and TMMLU+ as evaluation datasets, the system employs BGE-M3 for dense vector retrieval to obtain highly relevant search results and BGE-reranker to reorder these results based on query relevance. The most pertinent retrieval outcomes serve as reference knowledge for a Large Language Model (LLM), enhancing its ability to answer questions and establishing a knowledge retrieval system grounded in generative AI. The system's effectiveness is assessed through a two-stage evaluation: automatic and assisted performance evaluations. The automatic evaluation calculates accuracy by comparing the model's auto-generated labels with ground truth answers, measuring performance under standardized conditions without human intervention. The assisted performance evaluation involves 20 finance-related multiple-choice questions answered by 20 participants without financial backgrounds. Initially, participants answer independently. Later, they receive system-generated reference information to assist in answering, examining whether the system improves accuracy when assistance is provided. The main contributions of this research are: (1) Enhanced LLM Capability: By integrating BGE-M3 and BGE-reranker, the system retrieves and reorders highly relevant results, reduces hallucinations, and dynamically accesses authorized or public knowledge sources. (2) Improved Data Privacy: A customized RAG architecture enables local operation of the LLM, eliminating the need to send private data to external servers. This approach enhances data security, reduces reliance on commercial services, lowers operational costs, and mitigates privacy risks.
comment: 8 pages, 13 figures, 1 table
♻ ☆ Balancing Act: Prioritization Strategies for LLM-Designed Restless Bandit Rewards
LLMs are increasingly used to design reward functions based on human preferences in Reinforcement Learning (RL). We focus on LLM-designed rewards for Restless Multi-Armed Bandits, a framework for allocating limited resources among agents. In applications such as public health, this approach empowers grassroots health workers to tailor automated allocation decisions to community needs. In the presence of multiple agents, altering the reward function based on human preferences can impact subpopulations very differently, leading to complex tradeoffs and a multi-objective resource allocation problem. We are the first to present a principled method termed Social Choice Language Model for dealing with these tradeoffs for LLM-designed rewards for multiagent planners in general and restless bandits in particular. The novel part of our model is a transparent and configurable selection component, called an adjudicator, external to the LLM that controls complex tradeoffs via a user-selected social welfare function. Our experiments demonstrate that our model reliably selects more effective, aligned, and balanced reward functions compared to purely LLM-based approaches.
♻ ☆ Learning to Assist Humans without Inferring Rewards NeurIPS
Assistive agents should make humans' lives easier. Classically, such assistance is studied through the lens of inverse reinforcement learning, where an assistive agent (e.g., a chatbot, a robot) infers a human's intention and then selects actions to help the human reach that goal. This approach requires inferring intentions, which can be difficult in high-dimensional settings. We build upon prior work that studies assistance through the lens of empowerment: an assistive agent aims to maximize the influence of the human's actions such that they exert a greater control over the environmental outcomes and can solve tasks in fewer steps. We lift the major limitation of prior work in this area--scalability to high-dimensional settings--with contrastive successor representations. We formally prove that these representations estimate a similar notion of empowerment to that studied by prior work and provide a ready-made mechanism for optimizing it. Empirically, our proposed method outperforms prior methods on synthetic benchmarks, and scales to Overcooked, a cooperative game setting. Theoretically, our work connects ideas from information theory, neuroscience, and reinforcement learning, and charts a path for representations to play a critical role in solving assistive problems.
comment: Conference on Neural Information Processing Systems (NeurIPS), 2024
♻ ☆ TPIA: Towards Target-specific Prompt Injection Attack against Code-oriented Large Language Models
Recently, code-oriented large language models (Code LLMs) have been widely exploited to simplify and facilitate programming. With these tools, developers can easily generate the desired complete functional code based on incomplete code snippets and natural language prompts. Unfortunately, a few pioneering works revealed that these Code LLMs are vulnerable to backdoor and adversarial attacks. The former poisons the training data or model parameters, hijacking the LLMs to generate malicious code snippets when encountering the trigger. The latter crafts malicious adversarial input codes to reduce the quality of the generated codes. However, both attacks have some inherent limitations: backdoor attacks rely on the adversary's capability of controlling the model training process; adversarial attacks struggle with fulfilling specific malicious purposes. This paper presents a novel attack paradigm against Code LLMs, namely target-specific prompt injection attack (TPIA). TPIA generates non-functional perturbations containing the information of malicious instructions and inserts them into the victim's code context by spreading them into potentially used dependencies (e.g., packages or RAG's knowledge base). It induces the Code LLMs to generate attacker-specified malicious code snippets at the target location. In general, we compress the attacker-specified malicious objective into the perturbation by adversarial optimization based on greedy token search. We collect 13 representative malicious objectives to design 31 threat cases for three popular programming languages. We show that our TPIA can successfully attack three representative open-source Code LLMs (with an ASR of up to 97.9%) and two mainstream commercial Code LLM-integrated applications (with an ASR of over 90%) in all threat cases, using only a 12-token perturbation. Our work alerts a new practical threat of using Code LLMs.
♻ ☆ PsyDI: Towards a Personalized and Progressively In-depth Chatbot for Psychological Measurements
In the field of psychology, traditional assessment methods, such as standardized scales, are frequently critiqued for their static nature, lack of personalization, and reduced participant engagement, while comprehensive counseling evaluations are often inaccessible. The complexity of quantifying psychological traits further limits these methods. Despite advances with large language models (LLMs), many still depend on single-round Question-and-Answer interactions. To bridge this gap, we introduce PsyDI, a personalized and progressively in-depth chatbot designed for psychological measurements, exemplified by its application in the Myers-Briggs Type Indicator (MBTI) framework. PsyDI leverages user-related multi-modal information and engages in customized, multi-turn interactions to provide personalized, easily accessible measurements, while ensuring precise MBTI type determination. To address the challenge of unquantifiable psychological traits, we introduce a novel training paradigm that involves learning the ranking of proxy variables associated with these traits, culminating in a robust score model for MBTI measurements. The score model enables PsyDI to conduct comprehensive and precise measurements through multi-turn interactions within a unified estimation context. Through various experiments, we validate the efficacy of both the score model and the PsyDI pipeline, demonstrating its potential to serve as a general framework for psychological measurements. Furthermore, the online deployment of PsyDI has garnered substantial user engagement, with over 3,000 visits, resulting in the collection of numerous multi-turn dialogues annotated with MBTI types, which facilitates further research. The source code for the training and web service components is publicly available as a part of OpenDILab at: https://github.com/opendilab/PsyDI
comment: 29 pages, 15 figures
♻ ☆ Mitigating Overfitting in Graph Neural Networks via Feature and Hyperplane Perturbation
Graph neural networks (GNNs) are commonly used in semi-supervised settings. Previous research has primarily focused on finding appropriate graph filters (e.g. aggregation methods) to perform well on both homophilic and heterophilic graphs. While these methods are effective, they can still suffer from the sparsity of node features, where the initial data contain few non-zero elements. This can lead to overfitting in certain dimensions in the first projection matrix, as training samples may not cover the entire range of graph filters (hyperplanes). To address this, we propose a novel data augmentation strategy. Specifically, by flipping both the initial features and hyperplane, we create additional space for training, which leads to more precise updates of the learnable parameters and improved robustness for unseen features during inference. To the best of our knowledge, this is the first attempt to mitigate the overfitting caused by the initial features. Extensive experiments on real-world datasets show that our proposed technique increases node classification accuracy by up to 46.5% relatively.
♻ ☆ Collaborative Gym: A Framework for Enabling and Evaluating Human-Agent Collaboration
Recent advancements in language models (LMs) have sparked growing interest in developing LM agents. While fully autonomous agents could excel in many scenarios, numerous use cases inherently require them to collaborate with humans due to humans' latent preferences, domain expertise, or need for control. To facilitate the study of human-agent collaboration, we present Collaborative Gym (Co-Gym), a general framework enabling asynchronous, tripartite interaction among agents, humans, and task environments. We instantiate Co-Gym with three representative tasks in both simulated and real-world conditions, and propose an evaluation framework that assesses both the collaboration outcomes and processes. Our findings reveal that collaborative agents consistently outperform their fully autonomous counterparts in task performance within those delivered cases, achieving win rates of 86% in Travel Planning, 74% in Tabular Analysis, and 66% in Related Work when evaluated by real users. However, our study also highlights significant challenges in developing collaborative agents, requiring advancements in core aspects of intelligence -- communication capabilities, situational awareness, and balancing autonomy and human control.
comment: Preprint. Work in progress
♻ ☆ CMRxRecon2024: A Multi-Modality, Multi-View K-Space Dataset Boosting Universal Machine Learning for Accelerated Cardiac MRI
Cardiac magnetic resonance imaging (MRI) has emerged as a clinically gold-standard technique for diagnosing cardiac diseases, thanks to its ability to provide diverse information with multiple modalities and anatomical views. Accelerated cardiac MRI is highly expected to achieve time-efficient and patient-friendly imaging, and then advanced image reconstruction approaches are required to recover high-quality, clinically interpretable images from undersampled measurements. However, the lack of publicly available cardiac MRI k-space dataset in terms of both quantity and diversity has severely hindered substantial technological progress, particularly for data-driven artificial intelligence. Here, we provide a standardized, diverse, and high-quality CMRxRecon2024 dataset to facilitate the technical development, fair evaluation, and clinical transfer of cardiac MRI reconstruction approaches, towards promoting the universal frameworks that enable fast and robust reconstructions across different cardiac MRI protocols in clinical practice. To the best of our knowledge, the CMRxRecon2024 dataset is the largest and most protocal-diverse publicly available cardiac k-space dataset. It is acquired from 330 healthy volunteers, covering commonly used modalities, anatomical views, and acquisition trajectories in clinical cardiac MRI workflows. Besides, an open platform with tutorials, benchmarks, and data processing tools is provided to facilitate data usage, advanced method development, and fair performance evaluation.
comment: 23 pages, 3 figures, 2 tables
♻ ☆ MERaLiON-TextLLM: Cross-Lingual Understanding of Large Language Models in Chinese, Indonesian, Malay, and Singlish
Multilingual large language models (MLLMs) have shown impressive capabilities across a variety of languages. However, efficacy can differ greatly between different language families, especially for those with limited linguistic resources. This report presents MERaLiON-TextLLM, a series of open-source language models specifically tailored to improve understanding and generation in Chinese, Indonesian, Malay, and Singlish. The initial released model is built on Llama-3-8B-Base and refined through a meticulously crafted process of continued pre-training and weight merging. Our approach achieves performance improvements across benchmarks in these languages, exceeding the capabilities of the official Llama-3 models. We provide the model checkpoints as a resource to support further research and development in cross-lingual language understanding.
♻ ☆ Do LLMs Really Think Step-by-step In Implicit Reasoning?
It has been well-known that Chain-of-Thought can remarkably enhance LLMs' performance on complex tasks. However, because it also introduces slower inference speeds and higher computational costs, many researches have attempted to use implicit CoT, which does not need LLMs to explicitly generate the intermediate steps. However, the invisible reasoning process leaves us a doubt that, can implicit CoT really be equal to explicit CoT? Therefore, in this study, we address this question through experiments. We probe the information of intermediate steps from the model's hidden states when it is either trained or prompted to perform implicit CoT. The results surprisingly indicate that when prompted, LLMs hardly think about intermediate steps, suggesting they may just rely on experience rather than strict step-by-step reasoning. But when trained, they indeed calculate intermediate steps. Moreover, in both situations, we find the effect of using implicit CoT is susceptible to the format of the problem, reaffirming the current deficiency of implicit CoT.
comment: The code is in https://github.com/yuyijiong/if_step_by_step_implicit_CoT
♻ ☆ Measuring Diversity of Game Scenarios
This survey comprehensively reviews the multi-dimensionality of game scenario diversity, spotlighting the innovative use of procedural content generation and other fields as cornerstones for enriching player experiences through diverse game scenarios. By traversing a wide array of disciplines, from affective modeling and multi-agent systems to psychological studies, our research underscores the importance of diverse game scenarios in gameplay and education. Through a taxonomy of diversity metrics and evaluation methods, we aim to bridge the current gaps in literature and practice, offering insights into effective strategies for measuring and integrating diversity in game scenarios. Our analysis highlights the necessity for a unified taxonomy to aid developers and researchers in crafting more engaging and varied game worlds. This survey not only charts a path for future research in diverse game scenarios but also serves as a handbook for industry practitioners seeking to leverage diversity as a key component of game design and development.
♻ ☆ The surprising efficiency of temporal difference learning for rare event prediction NeurIPS 2024
We quantify the efficiency of temporal difference (TD) learning over the direct, or Monte Carlo (MC), estimator for policy evaluation in reinforcement learning, with an emphasis on estimation of quantities related to rare events. Policy evaluation is complicated in the rare event setting by the long timescale of the event and by the need for \emph{relative accuracy} in estimates of very small values. Specifically, we focus on least-squares TD (LSTD) prediction for finite state Markov chains, and show that LSTD can achieve relative accuracy far more efficiently than MC. We prove a central limit theorem for the LSTD estimator and upper bound the \emph{relative asymptotic variance} by simple quantities characterizing the connectivity of states relative to the transition probabilities between them. Using this bound, we show that, even when both the timescale of the rare event and the relative accuracy of the MC estimator are exponentially large in the number of states, LSTD maintains a fixed level of relative accuracy with a total number of observed transitions of the Markov chain that is only \emph{polynomially} large in the number of states.
comment: Final camera-ready version published at NeurIPS 2024. Correct an assumption statement and typos, and change/add a few sentences from the last version
♻ ☆ MERaLiON-AudioLLM: Bridging Audio and Language with Large Language Models
We introduce MERaLiON-AudioLLM (Multimodal Empathetic Reasoning and Learning in One Network), the first speech-text model tailored for Singapore's multilingual and multicultural landscape. Developed under the National Large Language Models Funding Initiative, Singapore, MERaLiON-AudioLLM integrates advanced speech and text processing to address the diverse linguistic nuances of local accents and dialects, enhancing accessibility and usability in complex, multilingual environments. Our results demonstrate improvements in both speech recognition and task-specific understanding, positioning MERaLiON-AudioLLM as a pioneering solution for region specific AI applications. We envision this release to set a precedent for future models designed to address localised linguistic and cultural contexts in a global framework.
comment: https://huggingface.co/MERaLiON/MERaLiON-AudioLLM-Whisper-SEA-LION
♻ ☆ CrisisSense-LLM: Instruction Fine-Tuned Large Language Model for Multi-label Social Media Text Classification in Disaster Informatics
In the field of crisis/disaster informatics, social media is increasingly being used for improving situational awareness to inform response and relief efforts. Efficient and accurate text classification tools have been a focal area of investigation in crisis informatics. However, current methods mostly rely on single-label text classification models, which fails to capture different insights embedded in dynamic and multifaceted disaster-related social media data. This study introduces a novel approach to disaster text classification by enhancing a pre-trained Large Language Model (LLM) through instruction fine-tuning targeted for multi-label classification of disaster-related tweets. Our methodology involves creating a comprehensive instruction dataset from disaster-related tweets, which is then used to fine-tune an open-source LLM, thereby embedding it with disaster-specific knowledge. This fine-tuned model can classify multiple aspects of disaster-related information simultaneously, such as the type of event, informativeness, and involvement of human aid, significantly improving the utility of social media data for situational awareness in disasters. The results demonstrate that this approach enhances the categorization of critical information from social media posts, thereby facilitating a more effective deployment for situational awareness during emergencies. This research paves the way for more advanced, adaptable, and robust disaster management tools, leveraging the capabilities of LLMs to improve real-time situational awareness and response strategies in disaster scenarios.
comment: Relevant source code and data is available: https://github.com/KaiYin97/CrsisLLM
♻ ☆ Can ChatGPT Overcome Behavioral Biases in the Financial Sector? Classify-and-Rethink: Multi-Step Zero-Shot Reasoning in the Gold Investment
Large Language Models (LLMs) have achieved remarkable success recently, displaying exceptional capabilities in creating understandable and organized text. These LLMs have been utilized in diverse fields, such as clinical research, where domain-specific models like Med-Palm have achieved human-level performance. Recently, researchers have employed advanced prompt engineering to enhance the general reasoning ability of LLMs. Despite the remarkable success of zero-shot Chain-of-Thoughts (CoT) in solving general reasoning tasks, the potential of these methods still remains paid limited attention in the financial reasoning task.To address this issue, we explore multiple prompt strategies and incorporated semantic news information to improve LLMs' performance on financial reasoning tasks.To the best of our knowledge, we are the first to explore this important issue by applying ChatGPT to the gold investment.In this work, our aim is to investigate the financial reasoning capabilities of LLMs and their capacity to generate logical and persuasive investment opinions. We will use ChatGPT, one of the most powerful LLMs recently, and prompt engineering to achieve this goal. Our research will focus on understanding the ability of LLMs in sophisticated analysis and reasoning within the context of investment decision-making. Our study finds that ChatGPT with CoT prompt can provide more explainable predictions and overcome behavioral biases, which is crucial in finance-related tasks and can achieve higher investment returns.
♻ ☆ Smoothness Really Matters: A Simple Yet Effective Approach for Unsupervised Graph Domain Adaptation AAAI2025
Unsupervised Graph Domain Adaptation (UGDA) seeks to bridge distribution shifts between domains by transferring knowledge from labeled source graphs to given unlabeled target graphs. Existing UGDA methods primarily focus on aligning features in the latent space learned by graph neural networks (GNNs) across domains, often overlooking structural shifts, resulting in limited effectiveness when addressing structurally complex transfer scenarios. Given the sensitivity of GNNs to local structural features, even slight discrepancies between source and target graphs could lead to significant shifts in node embeddings, thereby reducing the effectiveness of knowledge transfer. To address this issue, we introduce a novel approach for UGDA called Target-Domain Structural Smoothing (TDSS). TDSS is a simple and effective method designed to perform structural smoothing directly on the target graph, thereby mitigating structural distribution shifts and ensuring the consistency of node representations. Specifically, by integrating smoothing techniques with neighborhood sampling, TDSS maintains the structural coherence of the target graph while mitigating the risk of over-smoothing. Our theoretical analysis shows that TDSS effectively reduces target risk by improving model smoothness. Empirical results on three real-world datasets demonstrate that TDSS outperforms recent state-of-the-art baselines, achieving significant improvements across six transfer scenarios. The code is available in https://github.com/cwei01/TDSS.
comment: 11 pages, Accpected by AAAI2025
♻ ☆ MVGT: A Multi-view Graph Transformer Based on Spatial Relations for EEG Emotion Recognition
Electroencephalography (EEG), a technique that records electrical activity from the scalp using electrodes, plays a vital role in affective computing. However, fully utilizing the multi-domain characteristics of EEG signals remains a significant challenge. Traditional single-perspective analyses often fail to capture the complex interplay of temporal, frequency, and spatial dimensions in EEG data. To address this, we introduce a multi-view graph transformer (MVGT) based on spatial relations that integrates information across three domains: temporal dynamics from continuous series, frequency features extracted from frequency bands, and inter-channel relationships captured through several spatial encodings. This comprehensive approach allows model to capture the nuanced properties inherent in EEG signals, enhancing its flexibility and representational power. Evaluation on publicly available datasets demonstrates that MVGT surpasses state-of-the-art methods in performance. The results highlight its ability to extract multi-domain information and effectively model inter-channel relationships, showcasing its potential for EEG-based emotion recognition tasks.
♻ ☆ DiffMesh: A Motion-aware Diffusion Framework for Human Mesh Recovery from Videos WACV 2025
Human mesh recovery (HMR) provides rich human body information for various real-world applications. While image-based HMR methods have achieved impressive results, they often struggle to recover humans in dynamic scenarios, leading to temporal inconsistencies and non-smooth 3D motion predictions due to the absence of human motion. In contrast, video-based approaches leverage temporal information to mitigate this issue. In this paper, we present DiffMesh, an innovative motion-aware Diffusion-like framework for video-based HMR. DiffMesh establishes a bridge between diffusion models and human motion, efficiently generating accurate and smooth output mesh sequences by incorporating human motion within the forward process and reverse process in the diffusion model. Extensive experiments are conducted on the widely used datasets (Human3.6M \cite{h36m_pami} and 3DPW \cite{pw3d2018}), which demonstrate the effectiveness and efficiency of our DiffMesh. Visual comparisons in real-world scenarios further highlight DiffMesh's suitability for practical applications.
comment: WACV 2025
♻ ☆ Surveying Attitudinal Alignment Between Large Language Models Vs. Humans Towards 17 Sustainable Development Goals
Large Language Models (LLMs) have emerged as potent tools for advancing the United Nations' Sustainable Development Goals (SDGs). However, the attitudinal disparities between LLMs and humans towards these goals can pose significant challenges. This study conducts a comprehensive review and analysis of the existing literature on the attitudes of LLMs towards the 17 SDGs, emphasizing the comparison between their attitudes and support for each goal and those of humans. We examine the potential disparities, primarily focusing on aspects such as understanding and emotions, cultural and regional differences, task objective variations, and factors considered in the decision-making process. These disparities arise from the underrepresentation and imbalance in LLM training data, historical biases, quality issues, lack of contextual understanding, and skewed ethical values reflected. The study also investigates the risks and harms that may arise from neglecting the attitudes of LLMs towards the SDGs, including the exacerbation of social inequalities, racial discrimination, environmental destruction, and resource wastage. To address these challenges, we propose strategies and recommendations to guide and regulate the application of LLMs, ensuring their alignment with the principles and goals of the SDGs, and therefore creating a more just, inclusive, and sustainable future.
♻ ☆ Towards Balanced Continual Multi-Modal Learning in Human Pose Estimation
3D human pose estimation (3D HPE) has emerged as a prominent research topic, particularly in the realm of RGB-based methods. However, RGB images are susceptible to limitations such as sensitivity to lighting conditions and potential user discomfort. Consequently, multi-modal sensing, which leverages non-intrusive sensors, is gaining increasing attention. Nevertheless, multi-modal 3D HPE still faces challenges, including modality imbalance and the imperative for continual learning. In this work, we introduce a novel balanced continual multi-modal learning method for 3D HPE, which harnesses the power of RGB, LiDAR, mmWave, and WiFi. Specifically, we propose a Shapley value-based contribution algorithm to quantify the contribution of each modality and identify modality imbalance. To address this imbalance, we employ a re-learning strategy. Furthermore, recognizing that raw data is prone to noise contamination, we develop a novel denoising continual learning approach. This approach incorporates a noise identification and separation module to mitigate the adverse effects of noise and collaborates with the balanced learning strategy to enhance optimization. Additionally, an adaptive EWC mechanism is employed to alleviate catastrophic forgetting. We conduct extensive experiments on the widely-adopted multi-modal dataset, MM-Fi, which demonstrate the superiority of our approach in boosting 3D pose estimation and mitigating catastrophic forgetting in complex scenarios. We will release our codes.
♻ ☆ PeFoMed: Parameter Efficient Fine-tuning of Multimodal Large Language Models for Medical Imaging
Multimodal large language models (MLLMs) represent an evolutionary expansion in the capabilities of traditional large language models, enabling them to tackle challenges that surpass the scope of purely text-based applications. It leverages the knowledge previously encoded within these language models, thereby enhancing their applicability and functionality in the reign of multimodal contexts. Recent works investigate the adaptation of MLLMs as a universal solution to address medical multi-modal problems as a generative task. In this paper, we propose a parameter efficient framework for fine-tuning MLLMs, specifically validated on medical visual question answering (Med-VQA) and medical report generation (MRG) tasks, using public benchmark datasets. We also introduce an evaluation metric using the 5-point Likert scale and its weighted average value to measure the quality of the generated reports for MRG tasks, where the scale ratings are labelled by both humans manually and the GPT-4 model. We further assess the consistency of performance metrics across traditional measures, GPT-4, and human ratings for both VQA and MRG tasks. The results indicate that semantic similarity assessments using GPT-4 align closely with human annotators and provide greater stability, yet they reveal a discrepancy when compared to conventional lexical similarity measurements. This questions the reliability of lexical similarity metrics for evaluating the performance of generative models in Med-VQA and report generation tasks. Besides, our fine-tuned model significantly outperforms GPT-4v. This indicates that without additional fine-tuning, multi-modal models like GPT-4v do not perform effectively on medical imaging tasks. The code will be available here: https://github.com/jinlHe/PeFoMed.
comment: 12 pages, 8 figures, 12 tables
♻ ☆ Federated Deep Subspace Clustering
This paper introduces FDSC, a private-protected subspace clustering (SC) approach with federated learning (FC) schema. In each client, there is a deep subspace clustering network accounting for grouping the isolated data, composed of a encode network, a self-expressive layer, and a decode network. FDSC is achieved by uploading the encode network to communicate with other clients in the server. Besides, FDSC is also enhanced by preserving the local neighborhood relationship in each client. With the effects of federated learning and locality preservation, the learned data features from the encoder are boosted so as to enhance the self-expressiveness learning and result in better clustering performance. Experiments test FDSC on public datasets and compare with other clustering methods, demonstrating the effectiveness of FDSC.
comment: 8pages,4 figures, 4 Tables
♻ ☆ VBIM-Net: Variational Born Iterative Network for Inverse Scattering Problems
Recently, studies have shown the potential of integrating field-type iterative methods with deep learning (DL) techniques in solving inverse scattering problems (ISPs). In this article, we propose a novel Variational Born Iterative Network, namely, VBIM-Net, to solve the full-wave ISPs with significantly improved structural rationality and inversion quality. The proposed VBIM-Net emulates the alternating updates of the total electric field and the contrast in the variational Born iterative method (VBIM) by multiple layers of subnetworks. We embed the analytical calculation of the contrast variation into each subnetwork, converting the scattered field residual into an approximate contrast variation and then enhancing it by a U-Net, thus avoiding the requirement of matched measurement dimension and grid resolution as in existing approaches. The total field and contrast of each layer's output is supervised in the loss function of VBIM-Net, imposing soft physical constraints on the variables in the subnetworks, which benefits the model's performance. In addition, we design a training scheme with extra noise to enhance the model's stability. Extensive numerical results on synthetic and experimental data both verify the inversion quality, generalization ability, and robustness of the proposed VBIM-Net. This work may provide some new inspiration for the design of efficient field-type DL schemes.
comment: Accepted by IEEE Transactions on Geoscience and Remote Sensing
♻ ☆ DynST: Dynamic Sparse Training for Resource-Constrained Spatio-Temporal Forecasting
The ever-increasing sensor service, though opening a precious path and providing a deluge of earth system data for deep-learning-oriented earth science, sadly introduce a daunting obstacle to their industrial level deployment. Concretely, earth science systems rely heavily on the extensive deployment of sensors, however, the data collection from sensors is constrained by complex geographical and social factors, making it challenging to achieve comprehensive coverage and uniform deployment. To alleviate the obstacle, traditional approaches to sensor deployment utilize specific algorithms to design and deploy sensors. These methods \textit{dynamically adjust the activation times of sensors to optimize the detection process across each sub-region}. Regrettably, formulating an activation strategy generally based on historical observations and geographic characteristics, which make the methods and resultant models were neither simple nor practical. Worse still, the complex technical design may ultimately lead to a model with weak generalizability. In this paper, we introduce for the first time the concept of spatio-temporal data dynamic sparse training and are committed to adaptively, dynamically filtering important sensor distributions. To our knowledge, this is the \textbf{first} proposal (\textit{termed DynST}) of an \textbf{industry-level} deployment optimization concept at the data level. However, due to the existence of the temporal dimension, pruning of spatio-temporal data may lead to conflicts at different timestamps. To achieve this goal, we employ dynamic merge technology, along with ingenious dimensional mapping to mitigate potential impacts caused by the temporal aspect. During the training process, DynST utilize iterative pruning and sparse training, repeatedly identifying and dynamically removing sensor perception areas that contribute the least to future predictions.
♻ ☆ Enhanced Masked Image Modeling to Avoid Model Collapse on Multi-modal MRI Datasets
Multi-modal magnetic resonance imaging (MRI) provides information of lesions for computer-aided diagnosis from different views. Deep learning algorithms are suitable for identifying specific anatomical structures, segmenting lesions, and classifying diseases. Manual labels are limited due to the high expense, which hinders further improvement of accuracy. Self-supervised learning, particularly masked image modeling (MIM), has shown promise in utilizing unlabeled data. However, we spot model collapse when applying MIM to multi-modal MRI datasets. The performance of downstream tasks does not see any improvement following the collapsed model. To solve model collapse, we analyze and address it in two types: complete collapse and dimensional collapse. We find complete collapse occurs because the collapsed loss value in multi-modal MRI datasets falls below the normally converged loss value. Based on this, the hybrid mask pattern (HMP) masking strategy is introduced to elevate the collapsed loss above the normally converged loss value and avoid complete collapse. Additionally, we reveal that dimensional collapse stems from insufficient feature uniformity in MIM. We mitigate dimensional collapse by introducing the pyramid barlow twins (PBT) module as an explicit regularization method. Overall, we construct the enhanced MIM (E-MIM) with HMP and PBT module to avoid model collapse multi-modal MRI. Experiments are conducted on three multi-modal MRI datasets to validate the effectiveness of our approach in preventing both types of model collapse. By preventing model collapse, the training of the model becomes more stable, resulting in a decent improvement in performance for segmentation and classification tasks. The code is available at https://github.com/LinxuanHan/E-MIM.
comment: This work has been submitted to the lEEE for possible publication. copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ Enhancing Graph Self-Supervised Learning with Graph Interplay
Graph self-supervised learning (GSSL) has emerged as a compelling framework for extracting informative representations from graph-structured data without extensive reliance on labeled inputs. In this study, we introduce Graph Interplay (GIP), an innovative and versatile approach that significantly enhances the performance equipped with various existing GSSL methods. To this end, GIP advocates direct graph-level communications by introducing random inter-graph edges within standard batches. Against GIP's simplicity, we further theoretically show that \textsc{GIP} essentially performs a principled manifold separation via combining inter-graph message passing and GSSL, bringing about more structured embedding manifolds and thus benefits a series of downstream tasks. Our empirical study demonstrates that GIP surpasses the performance of prevailing GSSL methods across multiple benchmarks by significant margins, highlighting its potential as a breakthrough approach. Besides, GIP can be readily integrated into a series of GSSL methods and consistently offers additional performance gain. This advancement not only amplifies the capability of GSSL but also potentially sets the stage for a novel graph learning paradigm in a broader sense.
comment: Due to potential implicit data leakage in our experimental setup, where the pretraining dataset was ordered by default labels, we withdraw this manuscript for further self-examination and rigorous validation
♻ ☆ CryoBench: Diverse and challenging datasets for the heterogeneity problem in cryo-EM NeurIPS 2024
Cryo-electron microscopy (cryo-EM) is a powerful technique for determining high-resolution 3D biomolecular structures from imaging data. Its unique ability to capture structural variability has spurred the development of heterogeneous reconstruction algorithms that can infer distributions of 3D structures from noisy, unlabeled imaging data. Despite the growing number of advanced methods, progress in the field is hindered by the lack of standardized benchmarks with ground truth information and reliable validation metrics. Here, we introduce CryoBench, a suite of datasets, metrics, and benchmarks for heterogeneous reconstruction in cryo-EM. CryoBench includes five datasets representing different sources of heterogeneity and degrees of difficulty. These include conformational heterogeneity generated from designed motions of antibody complexes or sampled from a molecular dynamics simulation, as well as compositional heterogeneity from mixtures of ribosome assembly states or 100 common complexes present in cells. We then analyze state-of-the-art heterogeneous reconstruction tools, including neural and non-neural methods, assess their sensitivity to noise, and propose new metrics for quantitative evaluation. We hope that CryoBench will be a foundational resource for accelerating algorithmic development and evaluation in the cryo-EM and machine learning communities. Project page: https://cryobench.cs.princeton.edu.
comment: Accepted by NeurIPS 2024 (Spotlight)
Robotics 30
☆ Distilling Multi-modal Large Language Models for Autonomous Driving
Autonomous driving demands safe motion planning, especially in critical "long-tail" scenarios. Recent end-to-end autonomous driving systems leverage large language models (LLMs) as planners to improve generalizability to rare events. However, using LLMs at test time introduces high computational costs. To address this, we propose DiMA, an end-to-end autonomous driving system that maintains the efficiency of an LLM-free (or vision-based) planner while leveraging the world knowledge of an LLM. DiMA distills the information from a multi-modal LLM to a vision-based end-to-end planner through a set of specially designed surrogate tasks. Under a joint training strategy, a scene encoder common to both networks produces structured representations that are semantically grounded as well as aligned to the final planning objective. Notably, the LLM is optional at inference, enabling robust planning without compromising on efficiency. Training with DiMA results in a 37% reduction in the L2 trajectory error and an 80% reduction in the collision rate of the vision-based planner, as well as a 44% trajectory error reduction in longtail scenarios. DiMA also achieves state-of-the-art performance on the nuScenes planning benchmark.
☆ FAST: Efficient Action Tokenization for Vision-Language-Action Models
Autoregressive sequence models, such as Transformer-based vision-language action (VLA) policies, can be tremendously effective for capturing complex and generalizable robotic behaviors. However, such models require us to choose a tokenization of our continuous action signals, which determines how the discrete symbols predicted by the model map to continuous robot actions. We find that current approaches for robot action tokenization, based on simple per-dimension, per-timestep binning schemes, typically perform poorly when learning dexterous skills from high-frequency robot data. To address this challenge, we propose a new compression-based tokenization scheme for robot actions, based on the discrete cosine transform. Our tokenization approach, Frequency-space Action Sequence Tokenization (FAST), enables us to train autoregressive VLAs for highly dexterous and high-frequency tasks where standard discretization methods fail completely. Based on FAST, we release FAST+, a universal robot action tokenizer, trained on 1M real robot action trajectories. It can be used as a black-box tokenizer for a wide range of robot action sequences, with diverse action spaces and control frequencies. Finally, we show that, when combined with the pi0 VLA, our method can scale to training on 10k hours of robot data and match the performance of diffusion VLAs, while reducing training time by up to 5x.
comment: Website: https://www.pi.website/research/fast
☆ FLOL: Fast Baselines for Real-World Low-Light Enhancement
Low-Light Image Enhancement (LLIE) is a key task in computational photography and imaging. The problem of enhancing images captured during night or in dark environments has been well-studied in the image signal processing literature. However, current deep learning-based solutions struggle with efficiency and robustness in real-world scenarios (e.g. scenes with noise, saturated pixels, bad illumination). We propose a lightweight neural network that combines image processing in the frequency and spatial domains. Our method, FLOL+, is one of the fastest models for this task, achieving state-of-the-art results on popular real scenes datasets such as LOL and LSRW. Moreover, we are able to process 1080p images under 12ms. Code and models at https://github.com/cidautai/FLOL
comment: Technical Report
☆ CoNav Chair: Design of a ROS-based Smart Wheelchair for Shared Control Navigation in the Built Environment
With the number of people with disabilities (PWD) increasing worldwide each year, the demand for mobility support to enable independent living and social integration is also growing. Wheelchairs commonly support the mobility of PWD in both indoor and outdoor environments. However, current powered wheelchairs (PWC) often fail to meet the needs of PWD, who may find it difficult to operate them. Furthermore, existing research on robotic wheelchairs typically focuses either on full autonomy or enhanced manual control, which can lead to reduced efficiency and user trust. To address these issues, this paper proposes a Robot Operating System (ROS)-based smart wheelchair, called CoNav Chair, that incorporates a shared control navigation algorithm and obstacle avoidance to support PWD while fostering efficiency and trust between the robot and the user. Our design consists of hardware and software components. Experimental results conducted in a typical indoor social environment demonstrate the performance and effectiveness of the smart wheelchair hardware and software design. This integrated design promotes trust and autonomy, which are crucial for the acceptance of assistive mobility technologies in the built environment.
comment: 8 pages, 9 figures
☆ Model Predictive Path Integral Docking of Fully Actuated Surface Vessel
Autonomous docking remains one of the most challenging maneuvers in marine robotics, requiring precise control and robust perception in confined spaces. This paper presents a novel approach integrating Model Predictive Path Integral(MPPI) control with real-time LiDAR-based dock detection for autonomous surface vessel docking. Our framework uniquely combines probabilistic trajectory optimization with a multiobjective cost function that simultaneously considers docking precision, safety constraints, and motion efficiency. The MPPI controller generates optimal trajectories by intelligently sampling control sequences and evaluating their costs based on dynamic clearance requirements, orientation alignment, and target position objectives. We introduce an adaptive dock detection pipeline that processes LiDAR point clouds to extract critical geometric features, enabling real-time updates of docking parameters. The proposed method is extensively validated in a physics-based simulation environment that incorporates realistic sensor noise, vessel dynamics, and environmental constraints. Results demonstrate successful docking from various initial positions while maintaining safe clearances and smooth motion characteristics.
comment: 6 pages, 6 figures, 1 table, UT2025 Conference, IEEE International Symposium on Underwater Technology 2025
☆ Monte Carlo Tree Search with Velocity Obstacles for safe and efficient motion planning in dynamic environments
Online motion planning is a challenging problem for intelligent robots moving in dense environments with dynamic obstacles, e.g., crowds. In this work, we propose a novel approach for optimal and safe online motion planning with minimal information about dynamic obstacles. Specifically, our approach requires only the current position of the obstacles and their maximum speed, but it does not need any information about their exact trajectories or dynamic model. The proposed methodology combines Monte Carlo Tree Search (MCTS), for online optimal planning via model simulations, with Velocity Obstacles (VO), for obstacle avoidance. We perform experiments in a cluttered simulated environment with walls, and up to 40 dynamic obstacles moving with random velocities and directions. With an ablation study, we show the key contribution of VO in scaling up the efficiency of MCTS, selecting the safest and most rewarding actions in the tree of simulations. Moreover, we show the superiority of our methodology with respect to state-of-the-art planners, including Non-linear Model Predictive Control (NMPC), in terms of improved collision rate, computational and task performance.
☆ Mesh2SLAM in VR: A Fast Geometry-Based SLAM Framework for Rapid Prototyping in Virtual Reality Applications
SLAM is a foundational technique with broad applications in robotics and AR/VR. SLAM simulations evaluate new concepts, but testing on resource-constrained devices, such as VR HMDs, faces challenges: high computational cost and restricted sensor data access. This work proposes a sparse framework using mesh geometry projections as features, which improves efficiency and circumvents direct sensor data access, advancing SLAM research as we demonstrate in VR and through numerical evaluation.
☆ Comparison of Various SLAM Systems for Mobile Robot in an Indoor Environment
This article presents a comparative analysis of a mobile robot trajectories computed by various ROS-based SLAM systems. For this reason we developed a prototype of a mobile robot with common sensors: 2D lidar, a monocular and ZED stereo cameras. Then we conducted experiments in a typical office environment and collected data from all sensors, running all tested SLAM systems based on the acquired dataset. We studied the following SLAM systems: (a) 2D lidar-based: GMapping, Hector SLAM, Cartographer; (b) monocular camera-based: Large Scale Direct monocular SLAM (LSD SLAM), ORB SLAM, Direct Sparse Odometry (DSO); and (c) stereo camera-based: ZEDfu, Real-Time Appearance-Based Mapping (RTAB map), ORB SLAM, Stereo Parallel Tracking and Mapping (S-PTAM). Since all SLAM methods were tested on the same dataset we compared results for different SLAM systems with appropriate metrics, demonstrating encouraging results for lidar-based Cartographer SLAM, Monocular ORB SLAM and Stereo RTAB Map methods.
comment: 6 pages, 6 figures
☆ Sensorimotor Control Strategies for Tactile Robotics
How are robots becoming smarter at interacting with their surroundings? Recent advances have reshaped how robots use tactile sensing to perceive and engage with the world. Tactile sensing is a game-changer, allowing robots to embed sensorimotor control strategies to interact with complex environments and skillfully handle heterogeneous objects. Such control frameworks plan contact-driven motions while staying responsive to sudden changes. We review the latest methods for building perception and control systems in tactile robotics while offering practical guidelines for their design and implementation. We also address key challenges to shape the future of intelligent robots.
comment: 39 pages, 8 figures, 1 table
☆ Real-Time Generation of Near-Minimum-Energy Trajectories via Constraint-Informed Residual Learning
Industrial robotics demands significant energy to operate, making energy-reduction methodologies increasingly important. Strategies for planning minimum-energy trajectories typically involve solving nonlinear optimal control problems (OCPs), which rarely cope with real-time requirements. In this paper, we propose a paradigm for generating near minimum-energy trajectories for manipulators by learning from optimal solutions. Our paradigm leverages a residual learning approach, which embeds boundary conditions while focusing on learning only the adjustments needed to steer a standard solution to an optimal one. Compared to a computationally expensive OCP-based planner, our paradigm achieves 87.3% of the performance near the training dataset and 50.8% far from the dataset, while being two to three orders of magnitude faster.
☆ Path Planning for a UAV Swarm Using Formation Teaching-Learning-Based Optimization
This work addresses the path planning problem for a group of unmanned aerial vehicles (UAVs) to maintain a desired formation during operation. Our approach formulates the problem as an optimization task by defining a set of fitness functions that not only ensure the formation but also include constraints for optimal and safe UAV operation. To optimize the fitness function and obtain a suboptimal path, we employ the teaching-learning-based optimization algorithm and then further enhance it with mechanisms such as mutation, elite strategy, and multi-subject combination. A number of simulations and experiments have been conducted to evaluate the proposed method. The results demonstrate that the algorithm successfully generates valid paths for the UAVs to fly in a triangular formation for an inspection task.
comment: in Proceedings of the 2025 International Conference on Energy, Infrastructure and Environmental Research (EIER2025)
☆ Robust UAV Path Planning with Obstacle Avoidance for Emergency Rescue
The unmanned aerial vehicles (UAVs) are efficient tools for diverse tasks such as electronic reconnaissance, agricultural operations and disaster relief. In the complex three-dimensional (3D) environments, the path planning with obstacle avoidance for UAVs is a significant issue for security assurance. In this paper, we construct a comprehensive 3D scenario with obstacles and no-fly zones for dynamic UAV trajectory. Moreover, a novel artificial potential field algorithm coupled with simulated annealing (APF-SA) is proposed to tackle the robust path planning problem. APF-SA modifies the attractive and repulsive potential functions and leverages simulated annealing to escape local minimum and converge to globally optimal solutions. Simulation results demonstrate that the effectiveness of APF-SA, enabling efficient autonomous path planning for UAVs with obstacle avoidance.
☆ RoboReflect: Robotic Reflective Reasoning for Grasping Ambiguous-Condition Objects
As robotic technology rapidly develops, robots are being employed in an increasing number of fields. However, due to the complexity of deployment environments or the prevalence of ambiguous-condition objects, the practical application of robotics still faces many challenges, leading to frequent errors. Traditional methods and some LLM-based approaches, although improved, still require substantial human intervention and struggle with autonomous error correction in complex scenarios.In this work, we propose RoboReflect, a novel framework leveraging large vision-language models (LVLMs) to enable self-reflection and autonomous error correction in robotic grasping tasks. RoboReflect allows robots to automatically adjust their strategies based on unsuccessful attempts until successful execution is achieved.The corrected strategies are saved in a memory for future task reference.We evaluate RoboReflect through extensive testing on eight common objects prone to ambiguous conditions of three categories.Our results demonstrate that RoboReflect not only outperforms existing grasp pose estimation methods like AnyGrasp and high-level action planning techniques using GPT-4V but also significantly enhances the robot's ability to adapt and correct errors independently. These findings underscore the critical importance of autonomous selfreflection in robotic systems while effectively addressing the challenges posed by ambiguous environments.
☆ Interoceptive Robots for Convergent Shared Control in Collaborative Construction Work
Building autonomous mobile robots (AMRs) with optimized efficiency and adaptive capabilities-able to respond to changing task demands and dynamic environments-is a strongly desired goal for advancing construction robotics. Such robots can play a critical role in enabling automation, reducing operational carbon footprints, and supporting modular construction processes. Inspired by the adaptive autonomy of living organisms, we introduce interoception, which centers on the robot's internal state representation, as a foundation for developing self-reflection and conscious learning to enable continual learning and adaptability in robotic agents. In this paper, we factorize internal state variables and mathematical properties as "cognitive dissonance" in shared control paradigms, where human interventions occasionally occur. We offer a new perspective on how interoception can help build adaptive motion planning in AMRs by integrating the legacy of heuristic costs from grid/graph-based algorithms with recent advances in neuroscience and reinforcement learning. Declarative and procedural knowledge extracted from human semantic inputs is encoded into a hypergraph model that overlaps with the spatial configuration of onsite layout for path planning. In addition, we design a velocity-replay module using an encoder-decoder architecture with few-shot learning to enable robots to replicate velocity profiles in contextualized scenarios for multi-robot synchronization and handover collaboration. These "cached" knowledge representations are demonstrated in simulated environments for multi-robot motion planning and stacking tasks. The insights from this study pave the way toward artificial general intelligence in AMRs, fostering their progression from complexity to competence in construction automation.
☆ ThinTact:Thin Vision-Based Tactile Sensor by Lensless Imaging
Vision-based tactile sensors have drawn increasing interest in the robotics community. However, traditional lens-based designs impose minimum thickness constraints on these sensors, limiting their applicability in space-restricted settings. In this paper, we propose ThinTact, a novel lensless vision-based tactile sensor with a sensing field of over 200 mm2 and a thickness of less than 10 mm.ThinTact utilizes the mask-based lensless imaging technique to map the contact information to CMOS signals. To ensure real-time tactile sensing, we propose a real-time lensless reconstruction algorithm that leverages a frequency-spatial-domain joint filter based on discrete cosine transform (DCT). This algorithm achieves computation significantly faster than existing optimization-based methods. Additionally, to improve the sensing quality, we develop a mask optimization method based on the generic algorithm and the corresponding system matrix calibration algorithm.We evaluate the performance of our proposed lensless reconstruction and tactile sensing through qualitative and quantitative experiments. Furthermore, we demonstrate ThinTact's practical applicability in diverse applications, including texture recognition and contact-rich object manipulation. The paper will appear in the IEEE Transactions on Robotics: https://ieeexplore.ieee.org/document/10842357. Video: https://youtu.be/YrOO9BDMAHo
comment: \c{opyright} 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
☆ Are Open-Vocabulary Models Ready for Detection of MEP Elements on Construction Sites
The construction industry has long explored robotics and computer vision, yet their deployment on construction sites remains very limited. These technologies have the potential to revolutionize traditional workflows by enhancing accuracy, efficiency, and safety in construction management. Ground robots equipped with advanced vision systems could automate tasks such as monitoring mechanical, electrical, and plumbing (MEP) systems. The present research evaluates the applicability of open-vocabulary vision-language models compared to fine-tuned, lightweight, closed-set object detectors for detecting MEP components using a mobile ground robotic platform. A dataset collected with cameras mounted on a ground robot was manually annotated and analyzed to compare model performance. The results demonstrate that, despite the versatility of vision-language models, fine-tuned lightweight models still largely outperform them in specialized environments and for domain-specific tasks.
comment: 4 pages, 3 figures
☆ Torque Responsive Metamaterials Enable High Payload Soft Robot Arms
Soft robots have struggled to support large forces and moments while also supporting their own weight against gravity. This limits their ability to reach certain configurations necessary for tasks such as inspection and pushing objects up. We have overcome this limitation by creating an electrically driven metamaterial soft arm using handed shearing auxetics (HSA) and bendable extendable torque resistant (BETR) shafts. These use the large force and torque capacity of HSAs and the nestable torque transmission of BETRs to create a strong soft arm. We found that the HSA arm was able to push 2.3 kg vertically and lift more than 600 g when positioned horizontally, supporting 0.33 Nm of torque at the base. The arm is able to move between waypoints while carrying the large payload and demonstrates consistent movement with path variance below 5 mm. The HSA arm's ability to perform active grasping with HSA grippers was also demonstrated, requiring 20 N of pull force to dislodge the object. Finally, we test the arm in a pipe inspection task. The arm is able to locate all the defects while sliding against the inner surface of the pipe, demonstrating its compliance.
comment: 9 pages, 8 figures, currently under review
☆ GeoManip: Geometric Constraints as General Interfaces for Robot Manipulation
We present GeoManip, a framework to enable generalist robots to leverage essential conditions derived from object and part relationships, as geometric constraints, for robot manipulation. For example, cutting the carrot requires adhering to a geometric constraint: the blade of the knife should be perpendicular to the carrot's direction. By interpreting these constraints through symbolic language representations and translating them into low-level actions, GeoManip bridges the gap between natural language and robotic execution, enabling greater generalizability across diverse even unseen tasks, objects, and scenarios. Unlike vision-language-action models that require extensive training, operates training-free by utilizing large foundational models: a constraint generation module that predicts stage-specific geometric constraints and a geometry parser that identifies object parts involved in these constraints. A solver then optimizes trajectories to satisfy inferred constraints from task descriptions and the scene. Furthermore, GeoManip learns in-context and provides five appealing human-robot interaction features: on-the-fly policy adaptation, learning from human demonstrations, learning from failure cases, long-horizon action planning, and efficient data collection for imitation learning. Extensive evaluations on both simulations and real-world scenarios demonstrate GeoManip's state-of-the-art performance, with superior out-of-distribution generalization while avoiding costly model training.
comment: 32 pages, 13 figures
☆ SMPLest-X: Ultimate Scaling for Expressive Human Pose and Shape Estimation
Expressive human pose and shape estimation (EHPS) unifies body, hands, and face motion capture with numerous applications. Despite encouraging progress, current state-of-the-art methods focus on training innovative architectural designs on confined datasets. In this work, we investigate the impact of scaling up EHPS towards a family of generalist foundation models. 1) For data scaling, we perform a systematic investigation on 40 EHPS datasets, encompassing a wide range of scenarios that a model trained on any single dataset cannot handle. More importantly, capitalizing on insights obtained from the extensive benchmarking process, we optimize our training scheme and select datasets that lead to a significant leap in EHPS capabilities. Ultimately, we achieve diminishing returns at 10M training instances from diverse data sources. 2) For model scaling, we take advantage of vision transformers (up to ViT-Huge as the backbone) to study the scaling law of model sizes in EHPS. To exclude the influence of algorithmic design, we base our experiments on two minimalist architectures: SMPLer-X, which consists of an intermediate step for hand and face localization, and SMPLest-X, an even simpler version that reduces the network to its bare essentials and highlights significant advances in the capture of articulated hands. With big data and the large model, the foundation models exhibit strong performance across diverse test benchmarks and excellent transferability to even unseen environments. Moreover, our finetuning strategy turns the generalist into specialist models, allowing them to achieve further performance boosts. Notably, our foundation models consistently deliver state-of-the-art results on seven benchmarks such as AGORA, UBody, EgoBody, and our proposed SynHand dataset for comprehensive hand evaluation. (Code is available at: https://github.com/wqyin/SMPLest-X).
comment: An extension of SMPLer-X [arXiv:2309.17448]. Homepage: https://caizhongang.com/projects/SMPLer-X/
♻ ☆ Global SLAM in Visual-Inertial Systems with 5G Time-of-Arrival Integration
This paper presents a novel approach that integrates 5G Time of Arrival (ToA) measurements into ORB-SLAM3 to enable global localization and enhance mapping capabilities for indoor drone navigation. We extend ORB-SLAM3's optimization pipeline to jointly process ToA data from 5G base stations alongside visual and inertial measurements while estimating system biases. This integration transforms the inherently local SLAM estimates into globally referenced trajectories and effectively resolves scale ambiguity in monocular configurations. Our method is evaluated using five real-world indoor datasets collected with RGB-D cameras and inertial measurement units (IMUs), complemented by simulated 5G ToA measurements at 28 GHz and 78 GHz frequencies using MATLAB and QuaDRiGa. Extensive experiments across four SLAM configurations (RGB-D, RGB-D-Inertial, Monocular, and Monocular-Inertial) demonstrate that ToA integration enables consistent global positioning across all modes while significantly improving local accuracy in minimal sensor setups. Notably, ToA-enhanced monocular SLAM achieves superior local accuracy (6.3 cm average) compared to the RGB-D baseline (11.5 cm), and enables reliable operation of monocular-inertial SLAM in scenarios where the baseline system fails completely. While ToA integration offers limited local accuracy improvements for sensor-rich configurations like RGB-D SLAM, it consistently enables robust global localization.
♻ ☆ AeroHaptix: A Wearable Vibrotactile Feedback System for Enhancing Collision Avoidance in UAV Teleoperation
Haptic feedback enhances collision avoidance by providing directional obstacle information to operators during unmanned aerial vehicle (UAV) teleoperation. However, such feedback is often rendered via haptic joysticks, which are unfamiliar to UAV operators and limited to single-direction force feedback. Additionally, the direct coupling between the input device and the feedback method diminishes operators' sense of control and induces oscillatory movements. To overcome these limitations, we propose AeroHaptix, a wearable haptic feedback system that uses spatial vibrations to simultaneously communicate multiple obstacle directions to operators, without interfering with their input control. The layout of vibrotactile actuators was optimized via a perceptual study to eliminate perceptual biases and achieve uniform spatial coverage. A novel rendering algorithm, MultiCBF, extended control barrier functions to support multi-directional feedback. Our system evaluation showed that compared to a no-feedback condition, AeroHaptix effectively reduced the number of collisions and input disagreement. Furthermore, operators reported that AeroHaptix was more helpful than force feedback, with improved situational awareness and comparable workload.
♻ ☆ Learning Constraint Network from Demonstrations via Positive-Unlabeled Learning with Memory Replay
Planning for a wide range of real-world tasks necessitates to know and write all constraints. However, instances exist where these constraints are either unknown or challenging to specify accurately. A possible solution is to infer the unknown constraints from expert demonstration. The majority of prior works limit themselves to learning simple linear constraints, or require strong knowledge of the true constraint parameterization or environmental model. To mitigate these problems, this paper presents a positive-unlabeled (PU) learning approach to infer a continuous, arbitrary and possibly nonlinear, constraint from demonstration. From a PU learning view, We treat all data in demonstrations as positive (feasible) data, and learn a (sub)-optimal policy to generate high-reward-winning but potentially infeasible trajectories, which serve as unlabeled data containing both feasible and infeasible states. Under an assumption on data distribution, a feasible-infeasible classifier (i.e., constraint model) is learned from the two datasets through a postprocessing PU learning technique. The entire method employs an iterative framework alternating between updating the policy, which generates and selects higher-reward policies, and updating the constraint model. Additionally, a memory buffer is introduced to record and reuse samples from previous iterations to prevent forgetting. The effectiveness of the proposed method is validated in two Mujoco environments, successfully inferring continuous nonlinear constraints and outperforming a baseline method in terms of constraint accuracy and policy safety.
♻ ☆ Positive-Unlabeled Constraint Learning for Inferring Nonlinear Continuous Constraints Functions from Expert Demonstrations
Planning for diverse real-world robotic tasks necessitates to know and write all constraints. However, instances exist where these constraints are either unknown or challenging to specify accurately. A possible solution is to infer the unknown constraints from expert demonstration. This paper presents a novel two-step Positive-Unlabeled Constraint Learning (PUCL) algorithm to infer a continuous constraint function from demonstrations, without requiring prior knowledge of the true constraint parameterization or environmental model as existing works. We treat all data in demonstrations as positive (feasible) data, and learn a control policy to generate potentially infeasible trajectories, which serve as unlabeled data. The proposed two-step learning framework first identifies reliable infeasible data using a distance metric, and secondly learns a binary feasibility classifier (i.e., constraint function) from the feasible demonstrations and reliable infeasible data. The proposed method is flexible to learn complex-shaped constraint boundary and will not mistakenly classify demonstrations as infeasible as previous methods. The effectiveness of the proposed method is verified in four constrained environments, using a networked policy or a dynamical system policy. It successfully infers the continuous nonlinear constraints and outperforms other baseline methods in terms of constraint accuracy and policy safety. This work has been published in IEEE Robotics and Automation Letters (RA-L). Please refer to the final version at https://doi.org/10.1109/LRA.2024.3522756
♻ ☆ Humanoid Robot RHP Friends: Seamless Combination of Autonomous and Teleoperated Tasks in a Nursing Context
This paper describes RHP Friends, a social humanoid robot developed to enable assistive robotic deployments in human-coexisting environments. As a use-case application, we present its potential use in nursing by extending its capabilities to operate human devices and tools according to the task and by enabling remote assistance operations. To meet a wide variety of tasks and situations in environments designed by and for humans, we developed a system that seamlessly integrates the slim and lightweight robot and several technologies: locomanipulation, multi-contact motion, teleoperation, and object detection and tracking. We demonstrated the system's usage in a nursing application. The robot efficiently performed the daily task of patient transfer and a non-routine task, represented by a request to operate a circuit breaker. This demonstration, held at the 2023 International Robot Exhibition (IREX), conducted three times a day over three days.
comment: IEEE Robotics and Automation Magazine, In press
♻ ☆ Equivariant IMU Preintegration with Biases: a Galilean Group Approach
This letter proposes a new approach for Inertial Measurement Unit (IMU) preintegration, a fundamental building block that can be leveraged in different optimization-based Inertial Navigation System (INS) localization solutions. Inspired by recent advances in equivariant theory applied to biased INSs, we derive a discrete-time formulation of the IMU preintegration on ${\mathbf{Gal}(3) \ltimes \mathfrak{gal}(3)}$, the left-trivialization of the tangent group of the Galilean group $\mathbf{Gal}(3)$. We define a novel preintegration error that geometrically couples the navigation states and the bias leading to lower linearization error. Our method improves in consistency compared to existing preintegration approaches which treat IMU biases as a separate state-space. Extensive validation against state-of-the-art methods, both in simulation and with real-world IMU data, implementation in the Lie++ library, and open-source code are provided.
♻ ☆ PO-GVINS: Tightly Coupled GNSS-Visual-Inertial Integration with Pose-Only Representation
Accurate and reliable positioning is crucial for perception, decision-making, and other high-level applications in autonomous driving, unmanned aerial vehicles, and intelligent robots. Given the inherent limitations of standalone sensors, integrating heterogeneous sensors with complementary capabilities is one of the most effective approaches to achieving this goal. In this paper, we propose a filtering-based, tightly coupled global navigation satellite system (GNSS)-visual-inertial positioning framework with a pose-only formulation applied to the visual-inertial system (VINS), termed PO-GVINS. Specifically, multiple-view imaging used in current VINS requires a priori of 3D feature, then jointly estimate camera poses and 3D feature position, which inevitably introduces linearization error of the feature as well as facing dimensional explosion. However, the pose-only (PO) formulation, which is demonstrated to be equivalent to the multiple-view imaging and has been applied in visual reconstruction, represent feature depth using two camera poses and thus 3D feature position is removed from state vector avoiding aforementioned difficulties. Inspired by this, we first apply PO formulation in our VINS, i.e., PO-VINS. GNSS raw measurements are then incorporated with integer ambiguity resolved to achieve accurate and drift-free estimation. Extensive experiments demonstrate that the proposed PO-VINS significantly outperforms the multi-state constrained Kalman filter (MSCKF). By incorporating GNSS measurements, PO-GVINS achieves accurate, drift-free state estimation, making it a robust solution for positioning in challenging environments.
♻ ☆ Autonomous Algorithm for Training Autonomous Vehicles with Minimal Human Intervention
Recent reinforcement learning (RL) algorithms have demonstrated impressive results in simulated driving environments. However, autonomous vehicles trained in simulation often struggle to work well in the real world due to the fidelity gap between simulated and real-world environments. While directly training real-world autonomous vehicles with RL algorithms is a promising approach to bypass the fidelity gap problem, it presents several challenges. One critical yet often overlooked challenge is the need to reset a driving environment between every episode. This reset process demands significant human intervention, leading to poor training efficiency in the real world. In this paper, we introduce a novel autonomous algorithm that enables off-the-shelf RL algorithms to train autonomous vehicles with minimal human intervention. Our algorithm reduces unnecessary human intervention by aborting episodes to prevent unsafe states and identifying informative initial states for subsequent episodes. The key idea behind identifying informative initial states is to estimate the expected amount of information that can be obtained from under-explored but reachable states. Our algorithm also revisits rule-based autonomous driving algorithms and highlights their benefits in safely returning an autonomous vehicle to initial states. To evaluate how much human intervention is required during training, we implement challenging urban driving tasks that require an autonomous vehicle to reset to initial states on its own. The experimental results show that our autonomous algorithm is task-agnostic and achieves competitive driving performance with much less human intervention than baselines.
comment: 8 pages, 6 figures, 2 tables, conference
♻ ☆ Gameplay Filters: Robust Zero-Shot Safety through Adversarial Imagination
Despite the impressive recent advances in learning-based robot control, ensuring robustness to out-of-distribution conditions remains an open challenge. Safety filters can, in principle, keep arbitrary control policies from incurring catastrophic failures by overriding unsafe actions, but existing solutions for complex (e.g., legged) robot dynamics do not span the full motion envelope and instead rely on local, reduced-order models. These filters tend to overly restrict agility and can still fail when perturbed away from nominal conditions. This paper presents the gameplay filter, a new class of predictive safety filter that continually plays out hypothetical matches between its simulation-trained safety strategy and a virtual adversary co-trained to invoke worst-case events and sim-to-real error, and precludes actions that would cause failures down the line. We demonstrate the scalability and robustness of the approach with a first-of-its-kind full-order safety filter for (36-D) quadrupedal dynamics. Physical experiments on two different quadruped platforms demonstrate the superior zero-shot effectiveness of the gameplay filter under large perturbations such as tugging and unmodeled terrain. Experiment videos and open-source software are available online: https://saferobotics.org/research/gameplay-filter
♻ ☆ The Dark Side of Rich Rewards: Understanding and Mitigating Noise in VLM Rewards
While Vision-Language Models (VLMs) are increasingly used to generate reward signals for training embodied agents to follow instructions, our research reveals that agents guided by VLM rewards often underperform compared to those employing only intrinsic (exploration-driven) rewards, contradicting expectations set by recent work. We hypothesize that false positive rewards -- instances where unintended trajectories are incorrectly rewarded -- are more detrimental than false negatives. Our analysis confirms this hypothesis, revealing that the widely used cosine similarity metric is prone to false positive reward estimates. To address this, we introduce BiMI ({Bi}nary {M}utual {I}nformation), a novel reward function designed to mitigate noise. BiMI significantly enhances learning efficiency across diverse and challenging embodied navigation environments. Our findings offer a nuanced understanding of how different types of reward noise impact agent learning and highlight the importance of addressing multimodal reward signal noise when training embodied agents
comment: 11 main body pages, 21 appendix pages
♻ ☆ Visual collective behaviors on spherical robots
The implementation of collective motion, traditionally, disregard the limited sensing capabilities of an individual, to instead assuming an omniscient perception of the environment. This study implements a visual flocking model in a ``robot-in-the-loop'' approach to reproduce these behaviors with a flock composed of 10 independent spherical robots. The model achieves robotic collective motion by only using panoramic visual information of each robot, such as retinal position, optical size and optic flow of the neighboring robots. We introduce a virtual anchor to confine the collective robotic movements so to avoid wall interactions. For the first time, a simple visual robot-in-the-loop approach succeed in reproducing several collective motion phases, in particular, swarming, and milling. Another milestone achieved with by this model is bridging the gap between simulation and physical experiments by demonstrating nearly identical behaviors in both environments with the same visual model. To conclude, we show that our minimal visual collective motion model is sufficient to recreate most collective behaviors on a robot-in-the-loop system that is scalable, behaves as numerical simulations predict and is easily comparable to traditional models.
comment: 26 pages, 16 figures, journal bioinspired and biomimetics
Multimedia 7
☆ Metric Learning with Progressive Self-Distillation for Audio-Visual Embedding Learning ICASSP 2025
Metric learning projects samples into an embedded space, where similarities and dissimilarities are quantified based on their learned representations. However, existing methods often rely on label-guided representation learning, where representations of different modalities, such as audio and visual data, are aligned based on annotated labels. This approach tends to underutilize latent complex features and potential relationships inherent in the distributions of audio and visual data that are not directly tied to the labels, resulting in suboptimal performance in audio-visual embedding learning. To address this issue, we propose a novel architecture that integrates cross-modal triplet loss with progressive self-distillation. Our method enhances representation learning by leveraging inherent distributions and dynamically refining soft audio-visual alignments -- probabilistic alignments between audio and visual data that capture the inherent relationships beyond explicit labels. Specifically, the model distills audio-visual distribution-based knowledge from annotated labels in a subset of each batch. This self-distilled knowledge is used t
comment: 5 pages, 3 figures, 2 tables. Accepted by ICASSP 2025
☆ PAL: Prompting Analytic Learning with Missing Modality for Multi-Modal Class-Incremental Learning
Multi-modal class-incremental learning (MMCIL) seeks to leverage multi-modal data, such as audio-visual and image-text pairs, thereby enabling models to learn continuously across a sequence of tasks while mitigating forgetting. While existing studies primarily focus on the integration and utilization of multi-modal information for MMCIL, a critical challenge remains: the issue of missing modalities during incremental learning phases. This oversight can exacerbate severe forgetting and significantly impair model performance. To bridge this gap, we propose PAL, a novel exemplar-free framework tailored to MMCIL under missing-modality scenarios. Concretely, we devise modality-specific prompts to compensate for missing information, facilitating the model to maintain a holistic representation of the data. On this foundation, we reformulate the MMCIL problem into a Recursive Least-Squares task, delivering an analytical linear solution. Building upon these, PAL not only alleviates the inherent under-fitting limitation in analytic learning but also preserves the holistic representation of missing-modality data, achieving superior performance with less forgetting across various multi-modal incremental scenarios. Extensive experiments demonstrate that PAL significantly outperforms competitive methods across various datasets, including UPMC-Food101 and N24News, showcasing its robustness towards modality absence and its anti-forgetting ability to maintain high incremental accuracy.
☆ LAVCap: LLM-based Audio-Visual Captioning using Optimal Transport ICASSP 2025
Automated audio captioning is a task that generates textual descriptions for audio content, and recent studies have explored using visual information to enhance captioning quality. However, current methods often fail to effectively fuse audio and visual data, missing important semantic cues from each modality. To address this, we introduce LAVCap, a large language model (LLM)-based audio-visual captioning framework that effectively integrates visual information with audio to improve audio captioning performance. LAVCap employs an optimal transport-based alignment loss to bridge the modality gap between audio and visual features, enabling more effective semantic extraction. Additionally, we propose an optimal transport attention module that enhances audio-visual fusion using an optimal transport assignment map. Combined with the optimal training strategy, experimental results demonstrate that each component of our framework is effective. LAVCap outperforms existing state-of-the-art methods on the AudioCaps dataset, without relying on large datasets or post-processing. Code is available at https://github.com/NAVER-INTEL-Co-Lab/gaudi-lavcap.
comment: 5 pages, 2 figures; Accepted to ICASSP 2025
☆ MagnetDB: A Longitudinal Torrent Discovery Dataset with IMDb-Matched Movies and TV Shows
BitTorrent remains a prominent channel for illicit distribution of copyrighted material, yet the supply side of such content remains understudied. We introduce MagnetDB, a longitudinal dataset of torrents discovered through the BitTorrent DHT between 2018 and 2024, containing more than 28.6 million torrents and metadata of more than 950 million files. While our primary focus is on enabling research based on the supply of pirated movies and TV shows, the dataset also encompasses other legitimate and illegitimate torrents. By applying IMDb-matching and annotation to movie and TV show torrents, MagnetDB facilitates detailed analyses of pirated content evolution in the BitTorrent network. Researchers can leverage MagnetDB to examine distribution trends, subcultural practices, and the gift economy within piracy ecosystems. Through its scale and temporal scope, MagnetDB presents a unique opportunity for investigating the broader dynamics of BitTorrent and advancing empirical knowledge on digital piracy.
☆ SMPLest-X: Ultimate Scaling for Expressive Human Pose and Shape Estimation
Expressive human pose and shape estimation (EHPS) unifies body, hands, and face motion capture with numerous applications. Despite encouraging progress, current state-of-the-art methods focus on training innovative architectural designs on confined datasets. In this work, we investigate the impact of scaling up EHPS towards a family of generalist foundation models. 1) For data scaling, we perform a systematic investigation on 40 EHPS datasets, encompassing a wide range of scenarios that a model trained on any single dataset cannot handle. More importantly, capitalizing on insights obtained from the extensive benchmarking process, we optimize our training scheme and select datasets that lead to a significant leap in EHPS capabilities. Ultimately, we achieve diminishing returns at 10M training instances from diverse data sources. 2) For model scaling, we take advantage of vision transformers (up to ViT-Huge as the backbone) to study the scaling law of model sizes in EHPS. To exclude the influence of algorithmic design, we base our experiments on two minimalist architectures: SMPLer-X, which consists of an intermediate step for hand and face localization, and SMPLest-X, an even simpler version that reduces the network to its bare essentials and highlights significant advances in the capture of articulated hands. With big data and the large model, the foundation models exhibit strong performance across diverse test benchmarks and excellent transferability to even unseen environments. Moreover, our finetuning strategy turns the generalist into specialist models, allowing them to achieve further performance boosts. Notably, our foundation models consistently deliver state-of-the-art results on seven benchmarks such as AGORA, UBody, EgoBody, and our proposed SynHand dataset for comprehensive hand evaluation. (Code is available at: https://github.com/wqyin/SMPLest-X).
comment: An extension of SMPLer-X [arXiv:2309.17448]. Homepage: https://caizhongang.com/projects/SMPLer-X/
♻ ☆ Frechet Music Distance: A Metric For Generative Symbolic Music Evaluation
In this paper we introduce the Frechet Music Distance (FMD), a novel evaluation metric for generative symbolic music models, inspired by the Frechet Inception Distance (FID) in computer vision and Frechet Audio Distance (FAD) in generative audio. FMD calculates the distance between distributions of reference and generated symbolic music embeddings, capturing abstract musical features. We validate FMD across several datasets and models. Results indicate that FMD effectively differentiates model quality, providing a domain-specific metric for evaluating symbolic music generation, and establishing a reproducible standard for future research in symbolic music modeling.
♻ ☆ DiffMesh: A Motion-aware Diffusion Framework for Human Mesh Recovery from Videos WACV 2025
Human mesh recovery (HMR) provides rich human body information for various real-world applications. While image-based HMR methods have achieved impressive results, they often struggle to recover humans in dynamic scenarios, leading to temporal inconsistencies and non-smooth 3D motion predictions due to the absence of human motion. In contrast, video-based approaches leverage temporal information to mitigate this issue. In this paper, we present DiffMesh, an innovative motion-aware Diffusion-like framework for video-based HMR. DiffMesh establishes a bridge between diffusion models and human motion, efficiently generating accurate and smooth output mesh sequences by incorporating human motion within the forward process and reverse process in the diffusion model. Extensive experiments are conducted on the widely used datasets (Human3.6M \cite{h36m_pami} and 3DPW \cite{pw3d2018}), which demonstrate the effectiveness and efficiency of our DiffMesh. Visual comparisons in real-world scenarios further highlight DiffMesh's suitability for practical applications.
comment: WACV 2025
Computer Vision and Pattern Recognition 125
☆ Ouroboros-Diffusion: Exploring Consistent Content Generation in Tuning-free Long Video Diffusion
The first-in-first-out (FIFO) video diffusion, built on a pre-trained text-to-video model, has recently emerged as an effective approach for tuning-free long video generation. This technique maintains a queue of video frames with progressively increasing noise, continuously producing clean frames at the queue's head while Gaussian noise is enqueued at the tail. However, FIFO-Diffusion often struggles to keep long-range temporal consistency in the generated videos due to the lack of correspondence modeling across frames. In this paper, we propose Ouroboros-Diffusion, a novel video denoising framework designed to enhance structural and content (subject) consistency, enabling the generation of consistent videos of arbitrary length. Specifically, we introduce a new latent sampling technique at the queue tail to improve structural consistency, ensuring perceptually smooth transitions among frames. To enhance subject consistency, we devise a Subject-Aware Cross-Frame Attention (SACFA) mechanism, which aligns subjects across frames within short segments to achieve better visual coherence. Furthermore, we introduce self-recurrent guidance. This technique leverages information from all previous cleaner frames at the front of the queue to guide the denoising of noisier frames at the end, fostering rich and contextual global information interaction. Extensive experiments of long video generation on the VBench benchmark demonstrate the superiority of our Ouroboros-Diffusion, particularly in terms of subject consistency, motion smoothness, and temporal consistency.
☆ Multimodal LLMs Can Reason about Aesthetics in Zero-Shot
We present the first study on how Multimodal LLMs' (MLLMs) reasoning ability shall be elicited to evaluate the aesthetics of artworks. To facilitate this investigation, we construct MM-StyleBench, a novel high-quality dataset for benchmarking artistic stylization. We then develop a principled method for human preference modeling and perform a systematic correlation analysis between MLLMs' responses and human preference. Our experiments reveal an inherent hallucination issue of MLLMs in art evaluation, associated with response subjectivity. ArtCoT is proposed, demonstrating that art-specific task decomposition and the use of concrete language boost MLLMs' reasoning ability for aesthetics. Our findings offer valuable insights into MLLMs for art and can benefit a wide range of downstream applications, such as style transfer and artistic image generation. Code available at https://github.com/songrise/MLLM4Art.
comment: WIP, Homepage https://github.com/songrise/MLLM4Art
☆ SimGen: A Diffusion-Based Framework for Simultaneous Surgical Image and Segmentation Mask Generation
Acquiring and annotating surgical data is often resource-intensive, ethical constraining, and requiring significant expert involvement. While generative AI models like text-to-image can alleviate data scarcity, incorporating spatial annotations, such as segmentation masks, is crucial for precision-driven surgical applications, simulation, and education. This study introduces both a novel task and method, SimGen, for Simultaneous Image and Mask Generation. SimGen is a diffusion model based on the DDPM framework and Residual U-Net, designed to jointly generate high-fidelity surgical images and their corresponding segmentation masks. The model leverages cross-correlation priors to capture dependencies between continuous image and discrete mask distributions. Additionally, a Canonical Fibonacci Lattice (CFL) is employed to enhance class separability and uniformity in the RGB space of the masks. SimGen delivers high-fidelity images and accurate segmentation masks, outperforming baselines across six public datasets assessed on image and semantic inception distance metrics. Ablation study shows that the CFL improves mask quality and spatial separation. Downstream experiments suggest generated image-mask pairs are usable if regulations limit human data release for research. This work offers a cost-effective solution for generating paired surgical images and complex labels, advancing surgical AI development by reducing the need for expensive manual annotations.
comment: 12 pages, 17 figures, 4 tables, project page at https://camma-public.github.io/endogen/
☆ Vision Foundation Models for Computed Tomography
Foundation models (FMs) have shown transformative potential in radiology by performing diverse, complex tasks across imaging modalities. Here, we developed CT-FM, a large-scale 3D image-based pre-trained model designed explicitly for various radiological tasks. CT-FM was pre-trained using 148,000 computed tomography (CT) scans from the Imaging Data Commons through label-agnostic contrastive learning. We evaluated CT-FM across four categories of tasks, namely, whole-body and tumor segmentation, head CT triage, medical image retrieval, and semantic understanding, showing superior performance against state-of-the-art models. Beyond quantitative success, CT-FM demonstrated the ability to cluster regions anatomically and identify similar anatomical and structural concepts across scans. Furthermore, it remained robust across test-retest settings and indicated reasonable salient regions attached to its embeddings. This study demonstrates the value of large-scale medical imaging foundation models and by open-sourcing the model weights, code, and data, aims to support more adaptable, reliable, and interpretable AI solutions in radiology.
comment: 6 figures, followed by 9 Extended Data Figures and a Supplementary Information document
☆ RepVideo: Rethinking Cross-Layer Representation for Video Generation
Video generation has achieved remarkable progress with the introduction of diffusion models, which have significantly improved the quality of generated videos. However, recent research has primarily focused on scaling up model training, while offering limited insights into the direct impact of representations on the video generation process. In this paper, we initially investigate the characteristics of features in intermediate layers, finding substantial variations in attention maps across different layers. These variations lead to unstable semantic representations and contribute to cumulative differences between features, which ultimately reduce the similarity between adjacent frames and negatively affect temporal coherence. To address this, we propose RepVideo, an enhanced representation framework for text-to-video diffusion models. By accumulating features from neighboring layers to form enriched representations, this approach captures more stable semantic information. These enhanced representations are then used as inputs to the attention mechanism, thereby improving semantic expressiveness while ensuring feature consistency across adjacent frames. Extensive experiments demonstrate that our RepVideo not only significantly enhances the ability to generate accurate spatial appearances, such as capturing complex spatial relationships between multiple objects, but also improves temporal consistency in video generation.
comment: Project page: https://vchitect.github.io/RepVid-Webpage
☆ CityDreamer4D: Compositional Generative Model of Unbounded 4D Cities
3D scene generation has garnered growing attention in recent years and has made significant progress. Generating 4D cities is more challenging than 3D scenes due to the presence of structurally complex, visually diverse objects like buildings and vehicles, and heightened human sensitivity to distortions in urban environments. To tackle these issues, we propose CityDreamer4D, a compositional generative model specifically tailored for generating unbounded 4D cities. Our main insights are 1) 4D city generation should separate dynamic objects (e.g., vehicles) from static scenes (e.g., buildings and roads), and 2) all objects in the 4D scene should be composed of different types of neural fields for buildings, vehicles, and background stuff. Specifically, we propose Traffic Scenario Generator and Unbounded Layout Generator to produce dynamic traffic scenarios and static city layouts using a highly compact BEV representation. Objects in 4D cities are generated by combining stuff-oriented and instance-oriented neural fields for background stuff, buildings, and vehicles. To suit the distinct characteristics of background stuff and instances, the neural fields employ customized generative hash grids and periodic positional embeddings as scene parameterizations. Furthermore, we offer a comprehensive suite of datasets for city generation, including OSM, GoogleEarth, and CityTopia. The OSM dataset provides a variety of real-world city layouts, while the Google Earth and CityTopia datasets deliver large-scale, high-quality city imagery complete with 3D instance annotations. Leveraging its compositional design, CityDreamer4D supports a range of downstream applications, such as instance editing, city stylization, and urban simulation, while delivering state-of-the-art performance in generating realistic 4D cities.
☆ CityLoc: 6 DoF Localization of Text Descriptions in Large-Scale Scenes with Gaussian Representation
Localizing text descriptions in large-scale 3D scenes is inherently an ambiguous task. This nonetheless arises while describing general concepts, e.g. all traffic lights in a city. To facilitate reasoning based on such concepts, text localization in the form of distribution is required. In this paper, we generate the distribution of the camera poses conditioned upon the textual description. To facilitate such generation, we propose a diffusion-based architecture that conditionally diffuses the noisy 6DoF camera poses to their plausible locations. The conditional signals are derived from the text descriptions, using the pre-trained text encoders. The connection between text descriptions and pose distribution is established through pretrained Vision-Language-Model, i.e. CLIP. Furthermore, we demonstrate that the candidate poses for the distribution can be further refined by rendering potential poses using 3D Gaussian splatting, guiding incorrectly posed samples towards locations that better align with the textual description, through visual reasoning. We demonstrate the effectiveness of our method by comparing it with both standard retrieval methods and learning-based approaches. Our proposed method consistently outperforms these baselines across all five large-scale datasets. Our source code and dataset will be made publicly available.
☆ An analysis of data variation and bias in image-based dermatological datasets for machine learning classification
AI algorithms have become valuable in aiding professionals in healthcare. The increasing confidence obtained by these models is helpful in critical decision demands. In clinical dermatology, classification models can detect malignant lesions on patients' skin using only RGB images as input. However, most learning-based methods employ data acquired from dermoscopic datasets on training, which are large and validated by a gold standard. Clinical models aim to deal with classification on users' smartphone cameras that do not contain the corresponding resolution provided by dermoscopy. Also, clinical applications bring new challenges. It can contain captures from uncontrolled environments, skin tone variations, viewpoint changes, noises in data and labels, and unbalanced classes. A possible alternative would be to use transfer learning to deal with the clinical images. However, as the number of samples is low, it can cause degradations on the model's performance; the source distribution used in training differs from the test set. This work aims to evaluate the gap between dermoscopic and clinical samples and understand how the dataset variations impact training. It assesses the main differences between distributions that disturb the model's prediction. Finally, from experiments on different architectures, we argue how to combine the data from divergent distributions, decreasing the impact on the model's final accuracy.
comment: 10 pages, 1 figure
☆ Visual WetlandBirds Dataset: Bird Species Identification and Behavior Recognition in Videos
The current biodiversity loss crisis makes animal monitoring a relevant field of study. In light of this, data collected through monitoring can provide essential insights, and information for decision-making aimed at preserving global biodiversity. Despite the importance of such data, there is a notable scarcity of datasets featuring videos of birds, and none of the existing datasets offer detailed annotations of bird behaviors in video format. In response to this gap, our study introduces the first fine-grained video dataset specifically designed for bird behavior detection and species classification. This dataset addresses the need for comprehensive bird video datasets and provides detailed data on bird actions, facilitating the development of deep learning models to recognize these, similar to the advancements made in human action recognition. The proposed dataset comprises 178 videos recorded in Spanish wetlands, capturing 13 different bird species performing 7 distinct behavior classes. In addition, we also present baseline results using state of the art models on two tasks: bird behavior recognition and species classification.
☆ Learning Joint Denoising, Demosaicing, and Compression from the Raw Natural Image Noise Dataset
This paper introduces the Raw Natural Image Noise Dataset (RawNIND), a diverse collection of paired raw images designed to support the development of denoising models that generalize across sensors, image development workflows, and styles. Two denoising methods are proposed: one operates directly on raw Bayer data, leveraging computational efficiency, while the other processes linear RGB images for improved generalization to different sensors, with both preserving flexibility for subsequent development. Both methods outperform traditional approaches which rely on developed images. Additionally, the integration of denoising and compression at the raw data level significantly enhances rate-distortion performance and computational efficiency. These findings suggest a paradigm shift toward raw data workflows for efficient and flexible image processing.
☆ Empowering Agricultural Insights: RiceLeafBD - A Novel Dataset and Optimal Model Selection for Rice Leaf Disease Diagnosis through Transfer Learning Technique
The number of people living in this agricultural nation of ours, which is surrounded by lush greenery, is growing on a daily basis. As a result of this, the level of arable land is decreasing, as well as residential houses and industrial factories. The food crisis is becoming the main threat for us in the upcoming days. Because on the one hand, the population is increasing, and on the other hand, the amount of food crop production is decreasing due to the attack of diseases. Rice is one of the most significant cultivated crops since it provides food for more than half of the world's population. Bangladesh is dependent on rice (Oryza sativa) as a vital crop for its agriculture, but it faces a significant problem as a result of the ongoing decline in rice yield brought on by common diseases. Early disease detection is the main difficulty in rice crop cultivation. In this paper, we proposed our own dataset, which was collected from the Bangladesh field, and also applied deep learning and transfer learning models for the evaluation of the datasets. We elaborately explain our dataset and also give direction for further research work to serve society using this dataset. We applied a light CNN model and pre-trained InceptionNet-V2, EfficientNet-V2, and MobileNet-V2 models, which achieved 91.5% performance for the EfficientNet-V2 model of this work. The results obtained assaulted other models and even exceeded approaches that are considered to be part of the state of the art. It has been demonstrated by this study that it is possible to precisely and effectively identify diseases that affect rice leaves using this unbiased datasets. After analysis of the performance of different models, the proposed datasets are significant for the society for research work to provide solutions for decreasing rice leaf disease.
☆ Lights, Camera, Matching: The Role of Image Illumination in Fair Face Recognition
Facial brightness is a key image quality factor impacting face recognition accuracy differentials across demographic groups. In this work, we aim to decrease the accuracy gap between the similarity score distributions for Caucasian and African American female mated image pairs, as measured by d' between distributions. To balance brightness across demographic groups, we conduct three experiments, interpreting brightness in the face skin region either as median pixel value or as the distribution of pixel values. Balancing based on median brightness alone yields up to a 46.8% decrease in d', while balancing based on brightness distribution yields up to a 57.6% decrease. In all three cases, the similarity scores of the individual distributions improve, with mean scores maximally improving 5.9% for Caucasian females and 3.7% for African American females.
comment: 14 pages, 11 figures, Conference submission
☆ Multi-View Transformers for Airway-To-Lung Ratio Inference on Cardiac CT Scans: The C4R Study
The ratio of airway tree lumen to lung size (ALR), assessed at full inspiration on high resolution full-lung computed tomography (CT), is a major risk factor for chronic obstructive pulmonary disease (COPD). There is growing interest to infer ALR from cardiac CT images, which are widely available in epidemiological cohorts, to investigate the relationship of ALR to severe COVID-19 and post-acute sequelae of SARS-CoV-2 infection (PASC). Previously, cardiac scans included approximately 2/3 of the total lung volume with 5-6x greater slice thickness than high-resolution (HR) full-lung (FL) CT. In this study, we present a novel attention-based Multi-view Swin Transformer to infer FL ALR values from segmented cardiac CT scans. For the supervised training we exploit paired full-lung and cardiac CTs acquired in the Multi-Ethnic Study of Atherosclerosis (MESA). Our network significantly outperforms a proxy direct ALR inference on segmented cardiac CT scans and achieves accuracy and reproducibility comparable with a scan-rescan reproducibility of the FL ALR ground-truth.
comment: Accepted to appear in Proceedings of International Symposium on Biomedical Imaging (ISBI), 2025
☆ Enhanced Multi-Scale Cross-Attention for Person Image Generation ECCV2020
In this paper, we propose a novel cross-attention-based generative adversarial network (GAN) for the challenging person image generation task. Cross-attention is a novel and intuitive multi-modal fusion method in which an attention/correlation matrix is calculated between two feature maps of different modalities. Specifically, we propose the novel XingGAN (or CrossingGAN), which consists of two generation branches that capture the person's appearance and shape, respectively. Moreover, we propose two novel cross-attention blocks to effectively transfer and update the person's shape and appearance embeddings for mutual improvement. This has not been considered by any other existing GAN-based image generation work. To further learn the long-range correlations between different person poses at different scales and sub-regions, we propose two novel multi-scale cross-attention blocks. To tackle the issue of independent correlation computations within the cross-attention mechanism leading to noisy and ambiguous attention weights, which hinder performance improvements, we propose a module called enhanced attention (EA). Lastly, we introduce a novel densely connected co-attention module to fuse appearance and shape features at different stages effectively. Extensive experiments on two public datasets demonstrate that the proposed method outperforms current GAN-based methods and performs on par with diffusion-based methods. However, our method is significantly faster than diffusion-based methods in both training and inference.
comment: Accepted to TPAMI, an extended version of a paper published in ECCV2020. arXiv admin note: substantial text overlap with arXiv:2007.09278
☆ Feature-based One-For-All: A Universal Framework for Heterogeneous Knowledge Distillation
Knowledge distillation (KD) involves transferring knowledge from a pre-trained heavy teacher model to a lighter student model, thereby reducing the inference cost while maintaining comparable effectiveness. Prior KD techniques typically assume homogeneity between the teacher and student models. However, as technology advances, a wide variety of architectures have emerged, ranging from initial Convolutional Neural Networks (CNNs) to Vision Transformers (ViTs), and Multi-Level Perceptrons (MLPs). Consequently, developing a universal KD framework compatible with any architecture has become an important research topic. In this paper, we introduce a feature-based one-for-all (FOFA) KD framework to enable feature distillation across diverse architecture. Our framework comprises two key components. First, we design prompt tuning blocks that incorporate student feedback, allowing teacher features to adapt to the student model's learning process. Second, we propose region-aware attention to mitigate the view mismatch problem between heterogeneous architecture. By leveraging these two modules, effective distillation of intermediate features can be achieved across heterogeneous architectures. Extensive experiments on CIFAR, ImageNet, and COCO demonstrate the superiority of the proposed method.
☆ Generative Planning with 3D-vision Language Pre-training for End-to-End Autonomous Driving
Autonomous driving is a challenging task that requires perceiving and understanding the surrounding environment for safe trajectory planning. While existing vision-based end-to-end models have achieved promising results, these methods are still facing the challenges of vision understanding, decision reasoning and scene generalization. To solve these issues, a generative planning with 3D-vision language pre-training model named GPVL is proposed for end-to-end autonomous driving. The proposed paradigm has two significant aspects. On one hand, a 3D-vision language pre-training module is designed to bridge the gap between visual perception and linguistic understanding in the bird's eye view. On the other hand, a cross-modal language model is introduced to generate holistic driving decisions and fine-grained trajectories with perception and navigation information in an auto-regressive manner. Experiments on the challenging nuScenes dataset demonstrate that the proposed scheme achieves excellent performances compared with state-of-the-art methods. Besides, the proposed GPVL presents strong generalization ability and real-time potential when handling high-level commands in various scenarios. It is believed that the effective, robust and efficient performance of GPVL is crucial for the practical application of future autonomous driving systems. Code is available at https://github.com/ltp1995/GPVL
☆ Exploring Task-Level Optimal Prompts for Visual In-Context Learning
With the development of Vision Foundation Models (VFMs) in recent years, Visual In-Context Learning (VICL) has become a better choice compared to modifying models in most scenarios. Different from retraining or fine-tuning model, VICL does not require modifications to the model's weights or architecture, and only needs a prompt with demonstrations to teach VFM how to solve tasks. Currently, significant computational cost for finding optimal prompts for every test sample hinders the deployment of VICL, as determining which demonstrations to use for constructing prompts is very costly. In this paper, however, we find a counterintuitive phenomenon that most test samples actually achieve optimal performance under the same prompts, and searching for sample-level prompts only costs more time but results in completely identical prompts. Therefore, we propose task-level prompting to reduce the cost of searching for prompts during the inference stage and introduce two time-saving yet effective task-level prompt search strategies. Extensive experimental results show that our proposed method can identify near-optimal prompts and reach the best VICL performance with a minimal cost that prior work has never achieved.
☆ MANTA: Diffusion Mamba for Efficient and Effective Stochastic Long-Term Dense Anticipation
Our work addresses the problem of stochastic long-term dense anticipation. The goal of this task is to predict actions and their durations several minutes into the future based on provided video observations. Anticipation over extended horizons introduces high uncertainty, as a single observation can lead to multiple plausible future outcomes. To address this uncertainty, stochastic models are designed to predict several potential future action sequences. Recent work has further proposed to incorporate uncertainty modelling for observed frames by simultaneously predicting per-frame past and future actions in a unified manner. While such joint modelling of actions is beneficial, it requires long-range temporal capabilities to connect events across distant past and future time points. However, the previous work struggles to achieve such a long-range understanding due to its limited and/or sparse receptive field. To alleviate this issue, we propose a novel MANTA (MAmba for ANTicipation) network. Our model enables effective long-term temporal modelling even for very long sequences while maintaining linear complexity in sequence length. We demonstrate that our approach achieves state-of-the-art results on three datasets - Breakfast, 50Salads, and Assembly101 - while also significantly improving computational and memory efficiency.
☆ MMDocIR: Benchmarking Multi-Modal Retrieval for Long Documents
Multi-modal document retrieval is designed to identify and retrieve various forms of multi-modal content, such as figures, tables, charts, and layout information from extensive documents. Despite its significance, there is a notable lack of a robust benchmark to effectively evaluate the performance of systems in multi-modal document retrieval. To address this gap, this work introduces a new benchmark, named as MMDocIR, encompassing two distinct tasks: page-level and layout-level retrieval. The former focuses on localizing the most relevant pages within a long document, while the latter targets the detection of specific layouts, offering a more fine-grained granularity than whole-page analysis. A layout can refer to a variety of elements such as textual paragraphs, equations, figures, tables, or charts. The MMDocIR benchmark comprises a rich dataset featuring expertly annotated labels for 1,685 questions and bootstrapped labels for 173,843 questions, making it a pivotal resource for advancing multi-modal document retrieval for both training and evaluation. Through rigorous experiments, we reveal that (i) visual retrievers significantly outperform their text counterparts, (ii) MMDocIR train set can effectively benefit the training process of multi-modal document retrieval and (iii) text retrievers leveraging on VLM-text perform much better than those using OCR-text. These findings underscores the potential advantages of integrating visual elements for multi-modal document retrieval.
comment: https://huggingface.co/MMDocIR
☆ Boosting Diffusion Guidance via Learning Degradation-Aware Models for Blind Super Resolution WACV 2025
Recently, diffusion-based blind super-resolution (SR) methods have shown great ability to generate high-resolution images with abundant high-frequency detail, but the detail is often achieved at the expense of fidelity. Meanwhile, another line of research focusing on rectifying the reverse process of diffusion models (i.e., diffusion guidance), has demonstrated the power to generate high-fidelity results for non-blind SR. However, these methods rely on known degradation kernels, making them difficult to apply to blind SR. To address these issues, we introduce degradation-aware models that can be integrated into the diffusion guidance framework, eliminating the need to know degradation kernels. Additionally, we propose two novel techniques input perturbation and guidance scalar to further improve our performance. Extensive experimental results show that our proposed method has superior performance over state-of-the-art methods on blind SR benchmarks
comment: To appear in WACV 2025. Code is available at: https://github.com/ryanlu2240/Boosting-Diffusion-Guidance-via-Learning-Degradation-Aware-Models-for-Blind-Super-Resolution
☆ IDEA: Image Description Enhanced CLIP-Adapter
CLIP (Contrastive Language-Image Pre-training) has attained great success in pattern recognition and computer vision. Transferring CLIP to downstream tasks (e.g. zero- or few-shot classification) is a hot topic in multimodal learning. However, current studies primarily focus on either prompt learning for text or adapter tuning for vision, without fully exploiting the complementary information and correlations among image-text pairs. In this paper, we propose an Image Description Enhanced CLIP-Adapter (IDEA) method to adapt CLIP to few-shot image classification tasks. This method captures fine-grained features by leveraging both visual features and textual descriptions of images. IDEA is a training-free method for CLIP, and it can be comparable to or even exceeds state-of-the-art models on multiple tasks. Furthermore, we introduce Trainable-IDEA (T-IDEA), which extends IDEA by adding two lightweight learnable components (i.e., a projector and a learnable latent space), further enhancing the model's performance and achieving SOTA results on 11 datasets. As one important contribution, we employ the Llama model and design a comprehensive pipeline to generate textual descriptions for images of 11 datasets, resulting in a total of 1,637,795 image-text pairs, named "IMD-11". Our code and data are released at https://github.com/FourierAI/IDEA.
☆ Human Pose-Constrained UV Map Estimation
UV map estimation is used in computer vision for detailed analysis of human posture or activity. Previous methods assign pixels to body model vertices by comparing pixel descriptors independently, without enforcing global coherence or plausibility in the UV map. We propose Pose-Constrained Continuous Surface Embeddings (PC-CSE), which integrates estimated 2D human pose into the pixel-to-vertex assignment process. The pose provides global anatomical constraints, ensuring that UV maps remain coherent while preserving local precision. Evaluation on DensePose COCO demonstrates consistent improvement, regardless of the chosen 2D human pose model. Whole-body poses offer better constraints by incorporating additional details about the hands and feet. Conditioning UV maps with human pose reduces invalid mappings and enhances anatomical plausibility. In addition, we highlight inconsistencies in the ground-truth annotations.
☆ Multi-visual modality micro drone-based structural damage detection
Accurate detection and resilience of object detectors in structural damage detection are important in ensuring the continuous use of civil infrastructure. However, achieving robustness in object detectors remains a persistent challenge, impacting their ability to generalize effectively. This study proposes DetectorX, a robust framework for structural damage detection coupled with a micro drone. DetectorX addresses the challenges of object detector robustness by incorporating two innovative modules: a stem block and a spiral pooling technique. The stem block introduces a dynamic visual modality by leveraging the outputs of two Deep Convolutional Neural Network (DCNN) models. The framework employs the proposed event-based reward reinforcement learning to constrain the actions of a parent and child DCNN model leading to a reward. This results in the induction of two dynamic visual modalities alongside the Red, Green, and Blue (RGB) data. This enhancement significantly augments DetectorX's perception and adaptability in diverse environmental situations. Further, a spiral pooling technique, an online image augmentation method, strengthens the framework by increasing feature representations by concatenating spiraled and average/max pooled features. In three extensive experiments: (1) comparative and (2) robustness, which use the Pacific Earthquake Engineering Research Hub ImageNet dataset, and (3) field-experiment, DetectorX performed satisfactorily across varying metrics, including precision (0.88), recall (0.84), average precision (0.91), mean average precision (0.76), and mean average recall (0.73), compared to the competing detectors including You Only Look Once X-medium (YOLOX-m) and others. The study's findings indicate that DetectorX can provide satisfactory results and demonstrate resilience in challenging environments.
☆ Exploring ChatGPT for Face Presentation Attack Detection in Zero and Few-Shot in-Context Learning WACV
This study highlights the potential of ChatGPT (specifically GPT-4o) as a competitive alternative for Face Presentation Attack Detection (PAD), outperforming several PAD models, including commercial solutions, in specific scenarios. Our results show that GPT-4o demonstrates high consistency, particularly in few-shot in-context learning, where its performance improves as more examples are provided (reference data). We also observe that detailed prompts enable the model to provide scores reliably, a behavior not observed with concise prompts. Additionally, explanation-seeking prompts slightly enhance the model's performance by improving its interpretability. Remarkably, the model exhibits emergent reasoning capabilities, correctly predicting the attack type (print or replay) with high accuracy in few-shot scenarios, despite not being explicitly instructed to classify attack types. Despite these strengths, GPT-4o faces challenges in zero-shot tasks, where its performance is limited compared to specialized PAD systems. Experiments were conducted on a subset of the SOTERIA dataset, ensuring compliance with data privacy regulations by using only data from consenting individuals. These findings underscore GPT-4o's promise in PAD applications, laying the groundwork for future research to address broader data privacy concerns and improve cross-dataset generalization. Code available here: https://gitlab.idiap.ch/bob/bob.paper.wacv2025_chatgpt_face_pad
comment: Accepted in WACV workshop 2025
☆ Admitting Ignorance Helps the Video Question Answering Models to Answer
Significant progress has been made in the field of video question answering (VideoQA) thanks to deep learning and large-scale pretraining. Despite the presence of sophisticated model structures and powerful video-text foundation models, most existing methods focus solely on maximizing the correlation between answers and video-question pairs during training. We argue that these models often establish shortcuts, resulting in spurious correlations between questions and answers, especially when the alignment between video and text data is suboptimal. To address these spurious correlations, we propose a novel training framework in which the model is compelled to acknowledge its ignorance when presented with an intervened question, rather than making guesses solely based on superficial question-answer correlations. We introduce methodologies for intervening in questions, utilizing techniques such as displacement and perturbation, and design frameworks for the model to admit its lack of knowledge in both multi-choice VideoQA and open-ended settings. In practice, we integrate a state-of-the-art model into our framework to validate its effectiveness. The results clearly demonstrate that our framework can significantly enhance the performance of VideoQA models with minimal structural modifications.
☆ Few-Shot Learner Generalizes Across AI-Generated Image Detection
Current fake image detectors trained on large synthetic image datasets perform satisfactorily on limited studied generative models. However, they suffer a notable performance decline over unseen models. Besides, collecting adequate training data from online generative models is often expensive or infeasible. To overcome these issues, we propose Few-Shot Detector (FSD), a novel AI-generated image detector which learns a specialized metric space to effectively distinguish unseen fake images by utilizing very few samples. Experiments show FSD achieves state-of-the-art performance by $+7.4\%$ average ACC on GenImage dataset. More importantly, our method is better capable of capturing the intra-category common features in unseen images without further training.
comment: 11 pages, 5 figures
☆ $\texttt{InfoHier}$: Hierarchical Information Extraction via Encoding and Embedding
Analyzing large-scale datasets, especially involving complex and high-dimensional data like images, is particularly challenging. While self-supervised learning (SSL) has proven effective for learning representations from unlabelled data, it typically focuses on flat, non-hierarchical structures, missing the multi-level relationships present in many real-world datasets. Hierarchical clustering (HC) can uncover these relationships by organizing data into a tree-like structure, but it often relies on rigid similarity metrics that struggle to capture the complexity of diverse data types. To address these we envision $\texttt{InfoHier}$, a framework that combines SSL with HC to jointly learn robust latent representations and hierarchical structures. This approach leverages SSL to provide adaptive representations, enhancing HC's ability to capture complex patterns. Simultaneously, it integrates HC loss to refine SSL training, resulting in representations that are more attuned to the underlying information hierarchy. $\texttt{InfoHier}$ has the potential to improve the expressiveness and performance of both clustering and representation learning, offering significant benefits for data analysis, management, and information retrieval.
comment: 10 pages, 4 figures
Self-supervised Transformation Learning for Equivariant Representations NeurIPS 2024
Unsupervised representation learning has significantly advanced various machine learning tasks. In the computer vision domain, state-of-the-art approaches utilize transformations like random crop and color jitter to achieve invariant representations, embedding semantically the same inputs despite transformations. However, this can degrade performance in tasks requiring precise features, such as localization or flower classification. To address this, recent research incorporates equivariant representation learning, which captures transformation-sensitive information. However, current methods depend on transformation labels and thus struggle with interdependency and complex transformations. We propose Self-supervised Transformation Learning (STL), replacing transformation labels with transformation representations derived from image pairs. The proposed method ensures transformation representation is image-invariant and learns corresponding equivariant transformations, enhancing performance without increased batch complexity. We demonstrate the approach's effectiveness across diverse classification and detection tasks, outperforming existing methods in 7 out of 11 benchmarks and excelling in detection. By integrating complex transformations like AugMix, unusable by prior equivariant methods, this approach enhances performance across tasks, underscoring its adaptability and resilience. Additionally, its compatibility with various base models highlights its flexibility and broad applicability. The code is available at https://github.com/jaemyung-u/stl.
comment: 38th Conference on Neural Information Processing Systems (NeurIPS 2024)
☆ RealVVT: Towards Photorealistic Video Virtual Try-on via Spatio-Temporal Consistency
Virtual try-on has emerged as a pivotal task at the intersection of computer vision and fashion, aimed at digitally simulating how clothing items fit on the human body. Despite notable progress in single-image virtual try-on (VTO), current methodologies often struggle to preserve a consistent and authentic appearance of clothing across extended video sequences. This challenge arises from the complexities of capturing dynamic human pose and maintaining target clothing characteristics. We leverage pre-existing video foundation models to introduce RealVVT, a photoRealistic Video Virtual Try-on framework tailored to bolster stability and realism within dynamic video contexts. Our methodology encompasses a Clothing & Temporal Consistency strategy, an Agnostic-guided Attention Focus Loss mechanism to ensure spatial consistency, and a Pose-guided Long Video VTO technique adept at handling extended video sequences.Extensive experiments across various datasets confirms that our approach outperforms existing state-of-the-art models in both single-image and video VTO tasks, offering a viable solution for practical applications within the realms of fashion e-commerce and virtual fitting environments.
comment: 10 pages (8 pages main text, 2 pages references), 5 figures in the main text, and 4 pages supplementary materials with 3 additional figures
☆ FlexiClip: Locality-Preserving Free-Form Character Animation
Animating clipart images with seamless motion while maintaining visual fidelity and temporal coherence presents significant challenges. Existing methods, such as AniClipart, effectively model spatial deformations but often fail to ensure smooth temporal transitions, resulting in artifacts like abrupt motions and geometric distortions. Similarly, text-to-video (T2V) and image-to-video (I2V) models struggle to handle clipart due to the mismatch in statistical properties between natural video and clipart styles. This paper introduces FlexiClip, a novel approach designed to overcome these limitations by addressing the intertwined challenges of temporal consistency and geometric integrity. FlexiClip extends traditional B\'ezier curve-based trajectory modeling with key innovations: temporal Jacobians to correct motion dynamics incrementally, continuous-time modeling via probability flow ODEs (pfODEs) to mitigate temporal noise, and a flow matching loss inspired by GFlowNet principles to optimize smooth motion transitions. These enhancements ensure coherent animations across complex scenarios involving rapid movements and non-rigid deformations. Extensive experiments validate the effectiveness of FlexiClip in generating animations that are not only smooth and natural but also structurally consistent across diverse clipart types, including humans and animals. By integrating spatial and temporal modeling with pre-trained video diffusion models, FlexiClip sets a new standard for high-quality clipart animation, offering robust performance across a wide range of visual content. Project Page: https://creative-gen.github.io/flexiclip.github.io/
comment: 13 pages, 4 figures, 7 tables
☆ GS-LIVO: Real-Time LiDAR, Inertial, and Visual Multi-sensor Fused Odometry with Gaussian Mapping
In recent years, 3D Gaussian splatting (3D-GS) has emerged as a novel scene representation approach. However, existing vision-only 3D-GS methods often rely on hand-crafted heuristics for point-cloud densification and face challenges in handling occlusions and high GPU memory and computation consumption. LiDAR-Inertial-Visual (LIV) sensor configuration has demonstrated superior performance in localization and dense mapping by leveraging complementary sensing characteristics: rich texture information from cameras, precise geometric measurements from LiDAR, and high-frequency motion data from IMU. Inspired by this, we propose a novel real-time Gaussian-based simultaneous localization and mapping (SLAM) system. Our map system comprises a global Gaussian map and a sliding window of Gaussians, along with an IESKF-based odometry. The global Gaussian map consists of hash-indexed voxels organized in a recursive octree, effectively covering sparse spatial volumes while adapting to different levels of detail and scales. The Gaussian map is initialized through multi-sensor fusion and optimized with photometric gradients. Our system incrementally maintains a sliding window of Gaussians, significantly reducing GPU computation and memory consumption by only optimizing the map within the sliding window. Moreover, we implement a tightly coupled multi-sensor fusion odometry with an iterative error state Kalman filter (IESKF), leveraging real-time updating and rendering of the Gaussian map. Our system represents the first real-time Gaussian-based SLAM framework deployable on resource-constrained embedded systems, demonstrated on the NVIDIA Jetson Orin NX platform. The framework achieves real-time performance while maintaining robust multi-sensor fusion capabilities. All implementation algorithms, hardware designs, and CAD models will be publicly available.
☆ TimeFlow: Longitudinal Brain Image Registration and Aging Progression Analysis
Predicting future brain states is crucial for understanding healthy aging and neurodegenerative diseases. Longitudinal brain MRI registration, a cornerstone for such analyses, has long been limited by its inability to forecast future developments, reliance on extensive, dense longitudinal data, and the need to balance registration accuracy with temporal smoothness. In this work, we present \emph{TimeFlow}, a novel framework for longitudinal brain MRI registration that overcomes all these challenges. Leveraging a U-Net architecture with temporal conditioning inspired by diffusion models, TimeFlow enables accurate longitudinal registration and facilitates prospective analyses through future image prediction. Unlike traditional methods that depend on explicit smoothness regularizers and dense sequential data, TimeFlow achieves temporal consistency and continuity without these constraints. Experimental results highlight its superior performance in both future timepoint prediction and registration accuracy compared to state-of-the-art methods. Additionally, TimeFlow supports novel biological brain aging analyses, effectively differentiating neurodegenerative conditions from healthy aging. It eliminates the need for segmentation, thereby avoiding the challenges of non-trivial annotation and inconsistent segmentation errors. TimeFlow paves the way for accurate, data-efficient, and annotation-free prospective analyses of brain aging and chronic diseases.
☆ A Survey on Facial Image Privacy Preservation in Cloud-Based Services
Facial recognition models are increasingly employed by commercial enterprises, government agencies, and cloud service providers for identity verification, consumer services, and surveillance. These models are often trained using vast amounts of facial data processed and stored in cloud-based platforms, raising significant privacy concerns. Users' facial images may be exploited without their consent, leading to potential data breaches and misuse. This survey presents a comprehensive review of current methods aimed at preserving facial image privacy in cloud-based services. We categorize these methods into two primary approaches: image obfuscation-based protection and adversarial perturbation-based protection. We provide an in-depth analysis of both categories, offering qualitative and quantitative comparisons of their effectiveness. Additionally, we highlight unresolved challenges and propose future research directions to improve privacy preservation in cloud computing environments.
☆ Product of Gaussian Mixture Diffusion Model for non-linear MRI Inversion
Diffusion models have recently shown remarkable results in magnetic resonance imaging reconstruction. However, the employed networks typically are black-box estimators of the (smoothed) prior score with tens of millions of parameters, restricting interpretability and increasing reconstruction time. Furthermore, parallel imaging reconstruction algorithms either rely on off-line coil sensitivity estimation, which is prone to misalignment and restricting sampling trajectories, or perform per-coil reconstruction, making the computational cost proportional to the number of coils. To overcome this, we jointly reconstruct the image and the coil sensitivities using the lightweight, parameter-efficient, and interpretable product of Gaussian mixture diffusion model as an image prior and a classical smoothness priors on the coil sensitivities. The proposed method delivers promising results while allowing for fast inference and demonstrating robustness to contrast out-of-distribution data and sampling trajectories, comparable to classical variational penalties such as total variation. Finally, the probabilistic formulation allows the calculation of the posterior expectation and pixel-wise variance.
☆ BRIGHT-VO: Brightness-Guided Hybrid Transformer for Visual Odometry with Multi-modality Refinement Module
Visual odometry (VO) plays a crucial role in autonomous driving, robotic navigation, and other related tasks by estimating the position and orientation of a camera based on visual input. Significant progress has been made in data-driven VO methods, particularly those leveraging deep learning techniques to extract image features and estimate camera poses. However, these methods often struggle in low-light conditions because of the reduced visibility of features and the increased difficulty of matching keypoints. To address this limitation, we introduce BrightVO, a novel VO model based on Transformer architecture, which not only performs front-end visual feature extraction, but also incorporates a multi-modality refinement module in the back-end that integrates Inertial Measurement Unit (IMU) data. Using pose graph optimization, this module iteratively refines pose estimates to reduce errors and improve both accuracy and robustness. Furthermore, we create a synthetic low-light dataset, KiC4R, which includes a variety of lighting conditions to facilitate the training and evaluation of VO frameworks in challenging environments. Experimental results demonstrate that BrightVO achieves state-of-the-art performance on both the KiC4R dataset and the KITTI benchmarks. Specifically, it provides an average improvement of 20% in pose estimation accuracy in normal outdoor environments and 259% in low-light conditions, outperforming existing methods. For widespread use and further development, the research work is fully open-source at https://github.com/Anastasiawd/BrightVO.
comment: 9 pages, 7 figures
☆ StereoGen: High-quality Stereo Image Generation from a Single Image
State-of-the-art supervised stereo matching methods have achieved amazing results on various benchmarks. However, these data-driven methods suffer from generalization to real-world scenarios due to the lack of real-world annotated data. In this paper, we propose StereoGen, a novel pipeline for high-quality stereo image generation. This pipeline utilizes arbitrary single images as left images and pseudo disparities generated by a monocular depth estimation model to synthesize high-quality corresponding right images. Unlike previous methods that fill the occluded area in warped right images using random backgrounds or using convolutions to take nearby pixels selectively, we fine-tune a diffusion inpainting model to recover the background. Images generated by our model possess better details and undamaged semantic structures. Besides, we propose Training-free Confidence Generation and Adaptive Disparity Selection. The former suppresses the negative effect of harmful pseudo ground truth during stereo training, while the latter helps generate a wider disparity distribution and better synthetic images. Experiments show that models trained under our pipeline achieve state-of-the-art zero-shot generalization results among all published methods. The code will be available upon publication of the paper.
☆ Joint Learning of Depth and Appearance for Portrait Image Animation
2D portrait animation has experienced significant advancements in recent years. Much research has utilized the prior knowledge embedded in large generative diffusion models to enhance high-quality image manipulation. However, most methods only focus on generating RGB images as output, and the co-generation of consistent visual plus 3D output remains largely under-explored. In our work, we propose to jointly learn the visual appearance and depth simultaneously in a diffusion-based portrait image generator. Our method embraces the end-to-end diffusion paradigm and introduces a new architecture suitable for learning this conditional joint distribution, consisting of a reference network and a channel-expanded diffusion backbone. Once trained, our framework can be efficiently adapted to various downstream applications, such as facial depth-to-image and image-to-depth generation, portrait relighting, and audio-driven talking head animation with consistent 3D output.
☆ MonSter: Marry Monodepth to Stereo Unleashes Power
Stereo matching recovers depth from image correspondences. Existing methods struggle to handle ill-posed regions with limited matching cues, such as occlusions and textureless areas. To address this, we propose MonSter, a novel method that leverages the complementary strengths of monocular depth estimation and stereo matching. MonSter integrates monocular depth and stereo matching into a dual-branch architecture to iteratively improve each other. Confidence-based guidance adaptively selects reliable stereo cues for monodepth scale-shift recovery. The refined monodepth is in turn guides stereo effectively at ill-posed regions. Such iterative mutual enhancement enables MonSter to evolve monodepth priors from coarse object-level structures to pixel-level geometry, fully unlocking the potential of stereo matching. As shown in Fig.1, MonSter ranks 1st across five most commonly used leaderboards -- SceneFlow, KITTI 2012, KITTI 2015, Middlebury, and ETH3D. Achieving up to 49.5% improvements (Bad 1.0 on ETH3D) over the previous best method. Comprehensive analysis verifies the effectiveness of MonSter in ill-posed regions. In terms of zero-shot generalization, MonSter significantly and consistently outperforms state-of-the-art across the board. The code is publicly available at: https://github.com/Junda24/MonSter.
☆ Detecting Wildfire Flame and Smoke through Edge Computing using Transfer Learning Enhanced Deep Learning Models
Autonomous unmanned aerial vehicles (UAVs) integrated with edge computing capabilities empower real-time data processing directly on the device, dramatically reducing latency in critical scenarios such as wildfire detection. This study underscores Transfer Learning's (TL) significance in boosting the performance of object detectors for identifying wildfire smoke and flames, especially when trained on limited datasets, and investigates the impact TL has on edge computing metrics. With the latter focusing how TL-enhanced You Only Look Once (YOLO) models perform in terms of inference time, power usage, and energy consumption when using edge computing devices. This study utilizes the Aerial Fire and Smoke Essential (AFSE) dataset as the target, with the Flame and Smoke Detection Dataset (FASDD) and the Microsoft Common Objects in Context (COCO) dataset serving as source datasets. We explore a two-stage cascaded TL method, utilizing D-Fire or FASDD as initial stage target datasets and AFSE as the subsequent stage. Through fine-tuning, TL significantly enhances detection precision, achieving up to 79.2% mean Average Precision (mAP@0.5), reduces training time, and increases model generalizability across the AFSE dataset. However, cascaded TL yielded no notable improvements and TL alone did not benefit the edge computing metrics evaluated. Lastly, this work found that YOLOv5n remains a powerful model when lacking hardware acceleration, finding that YOLOv5n can process images nearly twice as fast as its newer counterpart, YOLO11n. Overall, the results affirm TL's role in augmenting the accuracy of object detectors while also illustrating that additional enhancements are needed to improve edge computing performance.
comment: 11 pages, 7 figures
☆ Self-Organizing Edge Computing Distribution Framework for Visual SLAM
Localization within a known environment is a crucial capability for mobile robots. Simultaneous Localization and Mapping (SLAM) is a prominent solution to this problem. SLAM is a framework that consists of a diverse set of computational tasks ranging from real-time tracking to computation-intensive map optimization. This combination can present a challenge for resource-limited mobile robots. Previously, edge-assisted SLAM methods have demonstrated promising real-time execution capabilities by offloading heavy computations while performing real-time tracking onboard. However, the common approach of utilizing a client-server architecture for offloading is sensitive to server and network failures. In this article, we propose a novel edge-assisted SLAM framework capable of self-organizing fully distributed SLAM execution across a network of devices or functioning on a single device without connectivity. The architecture consists of three layers and is designed to be device-agnostic, resilient to network failures, and minimally invasive to the core SLAM system. We have implemented and demonstrated the framework for monocular ORB SLAM3 and evaluated it in both fully distributed and standalone SLAM configurations against the ORB SLAM3. The experiment results demonstrate that the proposed design matches the accuracy and resource utilization of the monolithic approach while enabling collaborative execution.
comment: 8 pages, 5 figures
☆ Computerized Assessment of Motor Imitation for Distinguishing Autism in Video (CAMI-2DNet)
Motor imitation impairments are commonly reported in individuals with autism spectrum conditions (ASCs), suggesting that motor imitation could be used as a phenotype for addressing autism heterogeneity. Traditional methods for assessing motor imitation are subjective, labor-intensive, and require extensive human training. Modern Computerized Assessment of Motor Imitation (CAMI) methods, such as CAMI-3D for motion capture data and CAMI-2D for video data, are less subjective. However, they rely on labor-intensive data normalization and cleaning techniques, and human annotations for algorithm training. To address these challenges, we propose CAMI-2DNet, a scalable and interpretable deep learning-based approach to motor imitation assessment in video data, which eliminates the need for data normalization, cleaning and annotation. CAMI-2DNet uses an encoder-decoder architecture to map a video to a motion encoding that is disentangled from nuisance factors such as body shape and camera views. To learn a disentangled representation, we employ synthetic data generated by motion retargeting of virtual characters through the reshuffling of motion, body shape, and camera views, as well as real participant data. To automatically assess how well an individual imitates an actor, we compute a similarity score between their motion encodings, and use it to discriminate individuals with ASCs from neurotypical (NT) individuals. Our comparative analysis demonstrates that CAMI-2DNet has a strong correlation with human scores while outperforming CAMI-2D in discriminating ASC vs NT children. Moreover, CAMI-2DNet performs comparably to CAMI-3D while offering greater practicality by operating directly on video data and without the need for ad-hoc data normalization and human annotations.
comment: This work has been submitted to the IEEE for possible publication
☆ PACF: Prototype Augmented Compact Features for Improving Domain Adaptive Object Detection
In recent years, there has been significant advancement in object detection. However, applying off-the-shelf detectors to a new domain leads to significant performance drop, caused by the domain gap. These detectors exhibit higher-variance class-conditional distributions in the target domain than that in the source domain, along with mean shift. To address this problem, we propose the Prototype Augmented Compact Features (PACF) framework to regularize the distribution of intra-class features. Specifically, we provide an in-depth theoretical analysis on the lower bound of the target features-related likelihood and derive the prototype cross entropy loss to further calibrate the distribution of target RoI features. Furthermore, a mutual regularization strategy is designed to enable the linear and prototype-based classifiers to learn from each other, promoting feature compactness while enhancing discriminability. Thanks to this PACF framework, we have obtained a more compact cross-domain feature space, within which the variance of the target features' class-conditional distributions has significantly decreased, and the class-mean shift between the two domains has also been further reduced. The results on different adaptation settings are state-of-the-art, which demonstrate the board applicability and effectiveness of the proposed approach.
☆ Watermarking in Diffusion Model: Gaussian Shading with Exact Diffusion Inversion via Coupled Transformations (EDICT)
This paper introduces a novel approach to enhance the performance of Gaussian Shading, a prevalent watermarking technique, by integrating the Exact Diffusion Inversion via Coupled Transformations (EDICT) framework. While Gaussian Shading traditionally embeds watermarks in a noise latent space, followed by iterative denoising for image generation and noise addition for watermark recovery, its inversion process is not exact, leading to potential watermark distortion. We propose to leverage EDICT's ability to derive exact inverse mappings to refine this process. Our method involves duplicating the watermark-infused noisy latent and employing a reciprocal, alternating denoising and noising scheme between the two latents, facilitated by EDICT. This allows for a more precise reconstruction of both the image and the embedded watermark. Empirical evaluation on standard datasets demonstrates that our integrated approach yields a slight, yet statistically significant improvement in watermark recovery fidelity. These results highlight the potential of EDICT to enhance existing diffusion-based watermarking techniques by providing a more accurate and robust inversion mechanism. To the best of our knowledge, this is the first work to explore the synergy between EDICT and Gaussian Shading for digital watermarking, opening new avenues for research in robust and high-fidelity watermark embedding and extraction.
comment: 5 pages
☆ Image-to-Force Estimation for Soft Tissue Interaction in Robotic-Assisted Surgery Using Structured Light
For Minimally Invasive Surgical (MIS) robots, accurate haptic interaction force feedback is essential for ensuring the safety of interacting with soft tissue. However, most existing MIS robotic systems cannot facilitate direct measurement of the interaction force with hardware sensors due to space limitations. This letter introduces an effective vision-based scheme that utilizes a One-Shot structured light projection with a designed pattern on soft tissue coupled with haptic information processing through a trained image-to-force neural network. The images captured from the endoscopic stereo camera are analyzed to reconstruct high-resolution 3D point clouds for soft tissue deformation. Based on this, a modified PointNet-based force estimation method is proposed, which excels in representing the complex mechanical properties of soft tissue. Numerical force interaction experiments are conducted on three silicon materials with different stiffness. The results validate the effectiveness of the proposed scheme.
☆ Densely Connected Parameter-Efficient Tuning for Referring Image Segmentation AAAI2025
In the domain of computer vision, Parameter-Efficient Tuning (PET) is increasingly replacing the traditional paradigm of pre-training followed by full fine-tuning. PET is particularly favored for its effectiveness in large foundation models, as it streamlines transfer learning costs and optimizes hardware utilization. However, the current PET methods are mainly designed for single-modal optimization. While some pioneering studies have undertaken preliminary explorations, they still remain at the level of aligned encoders (e.g., CLIP) and lack exploration of misaligned encoders. These methods show sub-optimal performance with misaligned encoders, as they fail to effectively align the multimodal features during fine-tuning. In this paper, we introduce DETRIS, a parameter-efficient tuning framework designed to enhance low-rank visual feature propagation by establishing dense interconnections between each layer and all preceding layers, which enables effective cross-modal feature interaction and adaptation to misaligned encoders. We also suggest using text adapters to improve textual features. Our simple yet efficient approach greatly surpasses state-of-the-art methods with 0.9% to 1.8% backbone parameter updates, evaluated on challenging benchmarks. Our project is available at \url{https://github.com/jiaqihuang01/DETRIS}.
comment: Accepted by AAAI2025
☆ Scalable and High-Quality Neural Implicit Representation for 3D Reconstruction
Various SDF-based neural implicit surface reconstruction methods have been proposed recently, and have demonstrated remarkable modeling capabilities. However, due to the global nature and limited representation ability of a single network, existing methods still suffer from many drawbacks, such as limited accuracy and scale of the reconstruction. In this paper, we propose a versatile, scalable and high-quality neural implicit representation to address these issues. We integrate a divide-and-conquer approach into the neural SDF-based reconstruction. Specifically, we model the object or scene as a fusion of multiple independent local neural SDFs with overlapping regions. The construction of our representation involves three key steps: (1) constructing the distribution and overlap relationship of the local radiance fields based on object structure or data distribution, (2) relative pose registration for adjacent local SDFs, and (3) SDF blending. Thanks to the independent representation of each local region, our approach can not only achieve high-fidelity surface reconstruction, but also enable scalable scene reconstruction. Extensive experimental results demonstrate the effectiveness and practicality of our proposed method.
☆ GOTLoc: General Outdoor Text-based Localization Using Scene Graph Retrieval with OpenStreetMap
We propose GOTLoc, a robust localization method capable of operating even in outdoor environments where GPS signals are unavailable. The method achieves this robust localization by leveraging comparisons between scene graphs generated from text descriptions and maps. Existing text-based localization studies typically represent maps as point clouds and identify the most similar scenes by comparing embeddings of text and point cloud data. However, point cloud maps have limited scalability as it is impractical to pre-generate maps for all outdoor spaces. Furthermore, their large data size makes it challenging to store and utilize them directly on actual robots. To address these issues, GOTLoc leverages compact data structures, such as scene graphs, to store spatial information, enabling individual robots to carry and utilize large amounts of map data. Additionally, by utilizing publicly available map data, such as OpenStreetMap, which provides global information on outdoor spaces, we eliminate the need for additional effort to create custom map data. For performance evaluation, we utilized the KITTI360Pose dataset in conjunction with corresponding OpenStreetMap data to compare the proposed method with existing approaches. Our results demonstrate that the proposed method achieves accuracy comparable to algorithms relying on point cloud maps. Moreover, in city-scale tests, GOTLoc required significantly less storage compared to point cloud-based methods and completed overall processing within a few seconds, validating its applicability to real-world robotics. Our code is available at https://github.com/donghwijung/GOTLoc.
☆ MIAFEx: An Attention-based Feature Extraction Method for Medical Image Classification
Feature extraction techniques are crucial in medical image classification; however, classical feature extractors in addition to traditional machine learning classifiers often exhibit significant limitations in providing sufficient discriminative information for complex image sets. While Convolutional Neural Networks (CNNs) and Vision Transformer (ViT) have shown promise in feature extraction, they are prone to overfitting due to the inherent characteristics of medical imaging data, including small sample sizes or high intra-class variance. In this work, the Medical Image Attention-based Feature Extractor (MIAFEx) is proposed, a novel method that employs a learnable refinement mechanism to enhance the classification token within the Transformer encoder architecture. This mechanism adjusts the token based on learned weights, improving the extraction of salient features and enhancing the model's adaptability to the challenges presented by medical imaging data. The MIAFEx output features quality is compared against classical feature extractors using traditional and hybrid classifiers. Also, the performance of these features is compared against modern CNN and ViT models in classification tasks, demonstrating its superiority in accuracy and robustness across multiple complex classification medical imaging datasets. This advantage is particularly pronounced in scenarios with limited training data, where traditional and modern models often struggle to generalize effectively. The source code of this proposal can be found at https://github.com/Oscar-RamosS/Medical-Image-Attention-based-Feature-Extractor-MIAFEx
comment: In preparation for Journal Submission
☆ DynamicFace: High-Quality and Consistent Video Face Swapping using Composable 3D Facial Priors
Face swapping transfers the identity of a source face to a target face while retaining the attributes like expression, pose, hair, and background of the target face. Advanced face swapping methods have achieved attractive results. However, these methods often inadvertently transfer identity information from the target face, compromising expression-related details and accurate identity. We propose a novel method DynamicFace that leverages the power of diffusion model and plug-and-play temporal layers for video face swapping. First, we introduce four fine-grained face conditions using 3D facial priors. All conditions are designed to be disentangled from each other for precise and unique control. Then, we adopt Face Former and ReferenceNet for high-level and detailed identity injection. Through experiments on the FF++ dataset, we demonstrate that our method achieves state-of-the-art results in face swapping, showcasing superior image quality, identity preservation, and expression accuracy. Besides, our method could be easily transferred to video domain with temporal attention layer. Our code and results will be available on the project page: https://dynamic-face.github.io/
☆ The Devil is in Temporal Token: High Quality Video Reasoning Segmentation
Existing methods for Video Reasoning Segmentation rely heavily on a single special token to represent the object in the keyframe or the entire video, inadequately capturing spatial complexity and inter-frame motion. To overcome these challenges, we propose VRS-HQ, an end-to-end video reasoning segmentation approach that leverages Multimodal Large Language Models (MLLMs) to inject rich spatiotemporal features into hierarchical tokens.Our key innovations include a Temporal Dynamic Aggregation (TDA) and a Token-driven Keyframe Selection (TKS). Specifically, we design frame-level and temporal-level tokens that utilize MLLM's autoregressive learning to effectively capture both local and global information. Subsequently, we apply a similarity-based weighted fusion and frame selection strategy, then utilize SAM2 to perform keyframe segmentation and propagation. To enhance keyframe localization accuracy, the TKS filters keyframes based on SAM2's occlusion scores during inference. VRS-HQ achieves state-of-the-art performance on ReVOS, surpassing VISA by 5.9%/12.5%/9.1% in J&F scores across the three subsets. These results highlight the strong temporal reasoning and segmentation capabilities of our method. Code and model weights will be released at VRS-HQ.
☆ Comprehensive Subjective and Objective Evaluation Method for Text-generated Video
Recent text-to-video (T2V) technology advancements, as demonstrated by models such as Gen3, Pika, and Sora, have significantly broadened its applicability and popularity. This progress has created a growing demand for accurate quality assessment metrics to evaluate the perceptual quality of text-generated videos and optimize video generation models. However, assessing the quality of text-generated videos remains challenging due to the presence of highly complex distortions, such as unnatural actions and phenomena that defy human cognition. To address these challenges, we constructed a large-scale benchmark dataset for \textbf{T}ext-generated \textbf{V}ideo \textbf{eval}uation, \textbf{T2VEval-Bench}, comprising 148 textual words and 1,783 videos generated by 12 models. During the subjective evaluation, we collected five key scores: overall impression, video quality, aesthetic quality, realness, and text-video consistency. For objective evaluation, we developed the \textbf{T2VEval} model, which assesses videos across three branches: quality, authenticity, and consistency. Using an attention-based fusion module, T2VEval effectively integrates features from each branch and predicts scores with the aid of a large oracle model. Additionally, we implemented a progressive training strategy, enabling each branch to learn targeted knowledge while maintaining synergy with the others. Experimental results demonstrate that T2VEval achieves state-of-the-art performance across multiple metrics. The dataset and code will be open-sourced upon completion of the follow-up work.
☆ Multimodal Fake News Video Explanation Generation
Multi-modal explanation involves the assessment of the veracity of a variety of different content, and relies on multiple information modalities to comprehensively consider the relevance and consistency between modalities. Most existing fake news video detection methods focus on improving accuracy while ignoring the importance of providing explanations. In this paper, we propose a novel problem - Fake News Video Explanation (FNVE) - Given a multimodal news containing both video and caption text, we aim to generate natural language explanations to reveal the truth of predictions. To this end, we develop FakeNVE, a new dataset of explanations for truthfully multimodal posts, where each explanation is a natural language (English) sentence describing the attribution of a news thread. We benchmark FakeNVE by using a multimodal transformer-based architecture. Subsequently, a BART-based autoregressive decoder is used as the generator. Empirical results show compelling results for various baselines (applicable to FNVE) across multiple evaluation metrics. We also perform human evaluation on explanation generation, achieving high scores for both adequacy and fluency.
☆ Exploring the Efficacy of Meta-Learning: Unveiling Superior Data Diversity Utilization of MAML Over Pre-training
Currently, data and model size dominate the narrative in the training of super-large, powerful models. However, there has been a lack of exploration on the effect of other attributes of the training dataset on model performance. We hypothesize that dataset diversity can impact the performance of vision models. Our study shows positive correlations between test set accuracy and data diversity, providing an argument for furthering the research of dataset attributes beyond size. We analyzed pre-training and model-agnostic meta-learning methods on twelve popular visual datasets (e.g., Omniglot, CIFAR-FS, Aircraft) and five model configurations, including MAML variants with different numbers of inner gradient steps and supervised learning. We show moderate to strong positive correlations (R-squared: 0.15-0.42) between accuracy and data diversity and weaker but significant correlations (R-squared: ~0.2) between loss and diversity. These findings support our hypothesis and demonstrate a promising way for a deeper exploration of how formal data diversity influences model performance. This initial study highlights the potential of (Task2Vec) data diversity as a valuable measure in the rapidly evolving field of large-scale learning and emphasizes that understanding the dataset is key to building more powerful and generalizable models.
☆ Yuan: Yielding Unblemished Aesthetics Through A Unified Network for Visual Imperfections Removal in Generated Images
Generative AI presents transformative potential across various domains, from creative arts to scientific visualization. However, the utility of AI-generated imagery is often compromised by visual flaws, including anatomical inaccuracies, improper object placements, and misplaced textual elements. These imperfections pose significant challenges for practical applications. To overcome these limitations, we introduce \textit{Yuan}, a novel framework that autonomously corrects visual imperfections in text-to-image synthesis. \textit{Yuan} uniquely conditions on both the textual prompt and the segmented image, generating precise masks that identify areas in need of refinement without requiring manual intervention -- a common constraint in previous methodologies. Following the automated masking process, an advanced inpainting module seamlessly integrates contextually coherent content into the identified regions, preserving the integrity and fidelity of the original image and associated text prompts. Through extensive experimentation on publicly available datasets such as ImageNet100 and Stanford Dogs, along with a custom-generated dataset, \textit{Yuan} demonstrated superior performance in eliminating visual imperfections. Our approach consistently achieved higher scores in quantitative metrics, including NIQE, BRISQUE, and PI, alongside favorable qualitative evaluations. These results underscore \textit{Yuan}'s potential to significantly enhance the quality and applicability of AI-generated images across diverse fields.
☆ SuperSAM: Crafting a SAM Supernetwork via Structured Pruning and Unstructured Parameter Prioritization
Neural Architecture Search (NAS) is a powerful approach of automating the design of efficient neural architectures. In contrast to traditional NAS methods, recently proposed one-shot NAS methods prove to be more efficient in performing NAS. One-shot NAS works by generating a singular weight-sharing supernetwork that acts as a search space (container) of subnetworks. Despite its achievements, designing the one-shot search space remains a major challenge. In this work we propose a search space design strategy for Vision Transformer (ViT)-based architectures. In particular, we convert the Segment Anything Model (SAM) into a weight-sharing supernetwork called SuperSAM. Our approach involves automating the search space design via layer-wise structured pruning and parameter prioritization. While the structured pruning applies probabilistic removal of certain transformer layers, parameter prioritization performs weight reordering and slicing of MLP-blocks in the remaining layers. We train supernetworks on several datasets using the sandwich rule. For deployment, we enhance subnetwork discovery by utilizing a program autotuner to identify efficient subnetworks within the search space. The resulting subnetworks are 30-70% smaller in size compared to the original pre-trained SAM ViT-B, yet outperform the pretrained model. Our work introduces a new and effective method for ViT NAS search-space design.
☆ Unified Few-shot Crack Segmentation and its Precise 3D Automatic Measurement in Concrete Structures
Visual-Spatial Systems has become increasingly essential in concrete crack inspection. However, existing methods often lacks adaptability to diverse scenarios, exhibits limited robustness in image-based approaches, and struggles with curved or complex geometries. To address these limitations, an innovative framework for two-dimensional (2D) crack detection, three-dimensional (3D) reconstruction, and 3D automatic crack measurement was proposed by integrating computer vision technologies and multi-modal Simultaneous localization and mapping (SLAM) in this study. Firstly, building on a base DeepLabv3+ segmentation model, and incorporating specific refinements utilizing foundation model Segment Anything Model (SAM), we developed a crack segmentation method with strong generalization across unfamiliar scenarios, enabling the generation of precise 2D crack masks. To enhance the accuracy and robustness of 3D reconstruction, Light Detection and Ranging (LiDAR) point clouds were utilized together with image data and segmentation masks. By leveraging both image- and LiDAR-SLAM, we developed a multi-frame and multi-modal fusion framework that produces dense, colorized point clouds, effectively capturing crack semantics at a 3D real-world scale. Furthermore, the crack geometric attributions were measured automatically and directly within 3D dense point cloud space, surpassing the limitations of conventional 2D image-based measurements. This advancement makes the method suitable for structural components with curved and complex 3D geometries. Experimental results across various concrete structures highlight the significant improvements and unique advantages of the proposed method, demonstrating its effectiveness, accuracy, and robustness in real-world applications.
☆ Grounding Text-To-Image Diffusion Models For Controlled High-Quality Image Generation
Large-scale text-to-image (T2I) diffusion models have demonstrated an outstanding performance in synthesizing diverse high-quality visuals from natural language text captions. Multiple layout-to-image models have been developed to control the generation process by utilizing a broad array of layouts such as segmentation maps, edges, and human keypoints. In this work, we present ObjectDiffusion, a model that takes inspirations from the top cutting-edge image generative frameworks to seamlessly condition T2I models with new bounding boxes capabilities. Specifically, we make substantial modifications to the network architecture introduced in ContorlNet to integrate it with the condition processing and injection techniques proposed in GLIGEN. ObjectDiffusion is initialized with pretraining parameters to leverage the generation knowledge obtained from training on large-scale datasets. We fine-tune ObjectDiffusion on the COCO2017 training dataset and evaluate it on the COCO2017 validation dataset. Our model achieves an AP$_{50}$ of 46.6, an AR of 44.5, and a FID of 19.8 outperforming the current SOTA model trained on open-source datasets in all of the three metrics. ObjectDiffusion demonstrates a distinctive capability in synthesizing diverse, high-quality, high-fidelity images that seamlessly conform to the semantic and spatial control layout. Evaluated in qualitative and quantitative tests, ObjectDiffusion exhibits remarkable grounding abilities on closed-set and open-set settings across a wide variety of contexts. The qualitative assessment verifies the ability of ObjectDiffusion to generate multiple objects of different sizes and locations.
☆ Patch-aware Vector Quantized Codebook Learning for Unsupervised Visual Defect Detection ICTAI 2024
Unsupervised visual defect detection is critical in industrial applications, requiring a representation space that captures normal data features while detecting deviations. Achieving a balance between expressiveness and compactness is challenging; an overly expressive space risks inefficiency and mode collapse, impairing detection accuracy. We propose a novel approach using an enhanced VQ-VAE framework optimized for unsupervised defect detection. Our model introduces a patch-aware dynamic code assignment scheme, enabling context-sensitive code allocation to optimize spatial representation. This strategy enhances normal-defect distinction and improves detection accuracy during inference. Experiments on MVTecAD, BTAD, and MTSD datasets show our method achieves state-of-the-art performance.
comment: 7 pages, Accepted to 36th IEEE ICTAI 2024
☆ Cancer-Net PCa-Seg: Benchmarking Deep Learning Models for Prostate Cancer Segmentation Using Synthetic Correlated Diffusion Imaging
Prostate cancer (PCa) is the most prevalent cancer among men in the United States, accounting for nearly 300,000 cases, 29% of all diagnoses and 35,000 total deaths in 2024. Traditional screening methods such as prostate-specific antigen (PSA) testing and magnetic resonance imaging (MRI) have been pivotal in diagnosis, but have faced limitations in specificity and generalizability. In this paper, we explore the potential of enhancing PCa lesion segmentation using a novel MRI modality called synthetic correlated diffusion imaging (CDI$^s$). We employ several state-of-the-art deep learning models, including U-Net, SegResNet, Swin UNETR, Attention U-Net, and LightM-UNet, to segment PCa lesions from a 200 CDI$^s$ patient cohort. We find that SegResNet achieved superior segmentation performance with a Dice-Sorensen coefficient (DSC) of $76.68 \pm 0.8$. Notably, the Attention U-Net, while slightly less accurate (DSC $74.82 \pm 2.0$), offered a favorable balance between accuracy and computational efficiency. Our findings demonstrate the potential of deep learning models in improving PCa lesion segmentation using CDI$^s$ to enhance PCa management and clinical support.
comment: 8 pages, 2 figures, to be published in Studies in Computational Intelligence. This paper introduces Cancer-Net PCa-Seg, a comprehensive evaluation of deep learning models for prostate cancer segmentation using synthetic correlated diffusion imaging (CDI$^s$). We benchmark five state-of-the-art architectures: U-Net, SegResNet, Swin UNETR, Attention U-Net, and LightM-UNet
☆ Embodied Scene Understanding for Vision Language Models via MetaVQA
Vision Language Models (VLMs) demonstrate significant potential as embodied AI agents for various mobility applications. However, a standardized, closed-loop benchmark for evaluating their spatial reasoning and sequential decision-making capabilities is lacking. To address this, we present MetaVQA: a comprehensive benchmark designed to assess and enhance VLMs' understanding of spatial relationships and scene dynamics through Visual Question Answering (VQA) and closed-loop simulations. MetaVQA leverages Set-of-Mark prompting and top-down view ground-truth annotations from nuScenes and Waymo datasets to automatically generate extensive question-answer pairs based on diverse real-world traffic scenarios, ensuring object-centric and context-rich instructions. Our experiments show that fine-tuning VLMs with the MetaVQA dataset significantly improves their spatial reasoning and embodied scene comprehension in safety-critical simulations, evident not only in improved VQA accuracies but also in emerging safety-aware driving maneuvers. In addition, the learning demonstrates strong transferability from simulation to real-world observation. Code and data will be publicly available at https://metadriverse.github.io/metavqa .
comment: for the project webpage, see https://metadriverse.github.io/metavqa
☆ A Vessel Bifurcation Landmark Pair Dataset for Abdominal CT Deformable Image Registration (DIR) Validation
Deformable image registration (DIR) is an enabling technology in many diagnostic and therapeutic tasks. Despite this, DIR algorithms have limited clinical use, largely due to a lack of benchmark datasets for quality assurance during development. To support future algorithm development, here we introduce our first-of-its-kind abdominal CT DIR benchmark dataset, comprising large numbers of highly accurate landmark pairs on matching blood vessel bifurcations. Abdominal CT image pairs of 30 patients were acquired from several public repositories as well as the authors' institution with IRB approval. The two CTs of each pair were originally acquired for the same patient on different days. An image processing workflow was developed and applied to each image pair: 1) Abdominal organs were segmented with a deep learning model, and image intensity within organ masks was overwritten. 2) Matching image patches were manually identified between two CTs of each image pair 3) Vessel bifurcation landmarks were labeled on one image of each image patch pair. 4) Image patches were deformably registered, and landmarks were projected onto the second image. 5) Landmark pair locations were refined manually or with an automated process. This workflow resulted in 1895 total landmark pairs, or 63 per case on average. Estimates of the landmark pair accuracy using digital phantoms were 0.7+/-1.2mm. The data is published in Zenodo at https://doi.org/10.5281/zenodo.14362785. Instructions for use can be found at https://github.com/deshanyang/Abdominal-DIR-QA. This dataset is a first-of-its-kind for abdominal DIR validation. The number, accuracy, and distribution of landmark pairs will allow for robust validation of DIR algorithms with precision beyond what is currently available.
comment: 19 pages, 3 figures
☆ VCRScore: Image captioning metric based on V\&L Transformers, CLIP, and precision-recall
Image captioning has become an essential Vision & Language research task. It is about predicting the most accurate caption given a specific image or video. The research community has achieved impressive results by continuously proposing new models and approaches to improve the overall model's performance. Nevertheless, despite increasing proposals, the performance metrics used to measure their advances have remained practically untouched through the years. A probe of that, nowadays metrics like BLEU, METEOR, CIDEr, and ROUGE are still very used, aside from more sophisticated metrics such as BertScore and ClipScore. Hence, it is essential to adjust how are measure the advances, limitations, and scopes of the new image captioning proposals, as well as to adapt new metrics to these new advanced image captioning approaches. This work proposes a new evaluation metric for the image captioning problem. To do that, first, it was generated a human-labeled dataset to assess to which degree the captions correlate with the image's content. Taking these human scores as ground truth, we propose a new metric, and compare it with several well-known metrics, from classical to newer ones. Outperformed results were also found, and interesting insights were presented and discussed.
comment: 28 pages
☆ Few-Shot Adaptation of Training-Free Foundation Model for 3D Medical Image Segmentation
Vision foundation models have achieved remarkable progress across various image analysis tasks. In the image segmentation task, foundation models like the Segment Anything Model (SAM) enable generalizable zero-shot segmentation through user-provided prompts. However, SAM primarily trained on natural images, lacks the domain-specific expertise of medical imaging. This limitation poses challenges when applying SAM to medical image segmentation, including the need for extensive fine-tuning on specialized medical datasets and a dependency on manual prompts, which are both labor-intensive and require intervention from medical experts. This work introduces the Few-shot Adaptation of Training-frEe SAM (FATE-SAM), a novel method designed to adapt the advanced Segment Anything Model 2 (SAM2) for 3D medical image segmentation. FATE-SAM reassembles pre-trained modules of SAM2 to enable few-shot adaptation, leveraging a small number of support examples to capture anatomical knowledge and perform prompt-free segmentation, without requiring model fine-tuning. To handle the volumetric nature of medical images, we incorporate a Volumetric Consistency mechanism that enhances spatial coherence across 3D slices. We evaluate FATE-SAM on multiple medical imaging datasets and compare it with supervised learning methods, zero-shot SAM approaches, and fine-tuned medical SAM methods. Results show that FATE-SAM delivers robust and accurate segmentation while eliminating the need for large annotated datasets and expert intervention. FATE-SAM provides a practical, efficient solution for medical image segmentation, making it more accessible for clinical applications.
☆ Benchmarking Robustness of Contrastive Learning Models for Medical Image-Report Retrieval AAAI 2025
Medical images and reports offer invaluable insights into patient health. The heterogeneity and complexity of these data hinder effective analysis. To bridge this gap, we investigate contrastive learning models for cross-domain retrieval, which associates medical images with their corresponding clinical reports. This study benchmarks the robustness of four state-of-the-art contrastive learning models: CLIP, CXR-RePaiR, MedCLIP, and CXR-CLIP. We introduce an occlusion retrieval task to evaluate model performance under varying levels of image corruption. Our findings reveal that all evaluated models are highly sensitive to out-of-distribution data, as evidenced by the proportional decrease in performance with increasing occlusion levels. While MedCLIP exhibits slightly more robustness, its overall performance remains significantly behind CXR-CLIP and CXR-RePaiR. CLIP, trained on a general-purpose dataset, struggles with medical image-report retrieval, highlighting the importance of domain-specific training data. The evaluation of this work suggests that more effort needs to be spent on improving the robustness of these models. By addressing these limitations, we can develop more reliable cross-domain retrieval models for medical applications.
comment: This work is accepted to AAAI 2025 Workshop -- the 9th International Workshop on Health Intelligence
☆ Deep Self-Supervised Disturbance Mapping with the OPERA Sentinel-1 Radiometric Terrain Corrected SAR Backscatter Product
Mapping land surface disturbances supports disaster response, resource and ecosystem management, and climate adaptation efforts. Synthetic aperture radar (SAR) is an invaluable tool for disturbance mapping, providing consistent time-series images of the ground regardless of weather or illumination conditions. Despite SAR's potential for disturbance mapping, processing SAR data to an analysis-ready format requires expertise and significant compute resources, particularly for large-scale global analysis. In October 2023, NASA's Observational Products for End-Users from Remote Sensing Analysis (OPERA) project released the near-global Radiometric Terrain Corrected SAR backscatter from Sentinel-1 (RTC-S1) dataset, providing publicly available, analysis-ready SAR imagery. In this work, we utilize this new dataset to systematically analyze land surface disturbances. As labeling SAR data is often prohibitively time-consuming, we train a self-supervised vision transformer - which requires no labels to train - on OPERA RTC-S1 data to estimate a per-pixel distribution from the set of baseline imagery and assess disturbances when there is significant deviation from the modeled distribution. To test our model's capability and generality, we evaluate three different natural disasters - which represent high-intensity, abrupt disturbances - from three different regions of the world. Across events, our approach yields high quality delineations: F1 scores exceeding 0.6 and Areas Under the Precision-Recall Curve exceeding 0.65, consistently outperforming existing SAR disturbance methods. Our findings suggest that a self-supervised vision transformer is well-suited for global disturbance mapping and can be a valuable tool for operational, near-global disturbance monitoring, particularly when labeled data does not exist.
comment: 19 pages, 18 figures, 5 tables. Preprint. Submitted to JSTARS
☆ Deep Distance Map Regression Network with Shape-aware Loss for Imbalanced Medical Image Segmentation
Small object segmentation, like tumor segmentation, is a difficult and critical task in the field of medical image analysis. Although deep learning based methods have achieved promising performance, they are restricted to the use of binary segmentation mask. Inspired by the rigorous mapping between binary segmentation mask and distance map, we adopt distance map as a novel ground truth and employ a network to fulfill the computation of distance map. Specially, we propose a new segmentation framework that incorporates the existing binary segmentation network and a light weight regression network (dubbed as LR-Net). Thus, the LR-Net can convert the distance map computation into a regression task and leverage the rich information of distance maps. Additionally, we derive a shape-aware loss by employing distance maps as penalty map to infer the complete shape of an object. We evaluated our approach on MICCAI 2017 Liver Tumor Segmentation (LiTS) Challenge dataset and a clinical dataset. Experimental results show that our approach outperforms the classification-based methods as well as other existing state-of-the-arts.
comment: Conference
☆ Generative Medical Image Anonymization Based on Latent Code Projection and Optimization
Medical image anonymization aims to protect patient privacy by removing identifying information, while preserving the data utility to solve downstream tasks. In this paper, we address the medical image anonymization problem with a two-stage solution: latent code projection and optimization. In the projection stage, we design a streamlined encoder to project input images into a latent space and propose a co-training scheme to enhance the projection process. In the optimization stage, we refine the latent code using two deep loss functions designed to address the trade-off between identity protection and data utility dedicated to medical images. Through a comprehensive set of qualitative and quantitative experiments, we showcase the effectiveness of our approach on the MIMIC-CXR chest X-ray dataset by generating anonymized synthetic images that can serve as training set for detecting lung pathologies. Source codes are available at https://github.com/Huiyu-Li/GMIA.
comment: Conference
☆ Relation U-Net
Towards clinical interpretations, this paper presents a new ''output-with-confidence'' segmentation neural network with multiple input images and multiple output segmentation maps and their pairwise relations. A confidence score of the test image without ground-truth can be estimated from the difference among the estimated relation maps. We evaluate the method based on the widely used vanilla U-Net for segmentation and our new model is named Relation U-Net which can output segmentation maps of the input images as well as an estimated confidence score of the test image without ground-truth. Experimental results on four public datasets show that Relation U-Net can not only provide better accuracy than vanilla U-Net but also estimate a confidence score which is linearly correlated to the segmentation accuracy on test images.
comment: ISIB 2025
☆ Self Pre-training with Adaptive Mask Autoencoders for Variable-Contrast 3D Medical Imaging
The Masked Autoencoder (MAE) has recently demonstrated effectiveness in pre-training Vision Transformers (ViT) for analyzing natural images. By reconstructing complete images from partially masked inputs, the ViT encoder gathers contextual information to predict the missing regions. This capability to aggregate context is especially important in medical imaging, where anatomical structures are functionally and mechanically linked to surrounding regions. However, current methods do not consider variations in the number of input images, which is typically the case in real-world Magnetic Resonance (MR) studies. To address this limitation, we propose a 3D Adaptive Masked Autoencoders (AMAE) architecture that accommodates a variable number of 3D input contrasts per subject. A magnetic resonance imaging (MRI) dataset of 45,364 subjects was used for pretraining and a subset of 1648 training, 193 validation and 215 test subjects were used for finetuning. The performance demonstrates that self pre-training of this adaptive masked autoencoders can enhance the infarct segmentation performance by 2.8%-3.7% for ViT-based segmentation models.
comment: 5 pages, ISBI 2025 accepted
☆ Salient Information Preserving Adversarial Training Improves Clean and Robust Accuracy
In this work we introduce Salient Information Preserving Adversarial Training (SIP-AT), an intuitive method for relieving the robustness-accuracy trade-off incurred by traditional adversarial training. SIP-AT uses salient image regions to guide the adversarial training process in such a way that fragile features deemed meaningful by an annotator remain unperturbed during training, allowing models to learn highly predictive non-robust features without sacrificing overall robustness. This technique is compatible with both human-based and automatically generated salience estimates, allowing SIP-AT to be used as a part of human-driven model development without forcing SIP-AT to be reliant upon additional human data. We perform experiments across multiple datasets and architectures and demonstrate that SIP-AT is able to boost the clean accuracy of models while maintaining a high degree of robustness against attacks at multiple epsilon levels. We complement our central experiments with an observational study measuring the rate at which human subjects successfully identify perturbed images. This study helps build a more intuitive understanding of adversarial attack strength and demonstrates the heightened importance of low-epsilon robustness. Our results demonstrate the efficacy of SIP-AT and provide valuable insight into the risks posed by adversarial samples of various strengths.
☆ SHYI: Action Support for Contrastive Learning in High-Fidelity Text-to-Image Generation
In this project, we address the issue of infidelity in text-to-image generation, particularly for actions involving multiple objects. For this we build on top of the CONFORM framework which uses Contrastive Learning to improve the accuracy of the generated image for multiple objects. However the depiction of actions which involves multiple different object has still large room for improvement. To improve, we employ semantically hypergraphic contrastive adjacency learning, a comprehension of enhanced contrastive structure and "contrast but link" technique. We further amend Stable Diffusion's understanding of actions by InteractDiffusion. As evaluation metrics we use image-text similarity CLIP and TIFA. In addition, we conducted a user study. Our method shows promising results even with verbs that Stable Diffusion understands mediocrely. We then provide future directions by analyzing the results. Our codebase can be found on polybox under the link: https://polybox.ethz.ch/index.php/s/dJm3SWyRohUrFxn
comment: Main content 4 pages
♻ ☆ T2V-CompBench: A Comprehensive Benchmark for Compositional Text-to-video Generation
Text-to-video (T2V) generative models have advanced significantly, yet their ability to compose different objects, attributes, actions, and motions into a video remains unexplored. Previous text-to-video benchmarks also neglect this important ability for evaluation. In this work, we conduct the first systematic study on compositional text-to-video generation. We propose T2V-CompBench, the first benchmark tailored for compositional text-to-video generation. T2V-CompBench encompasses diverse aspects of compositionality, including consistent attribute binding, dynamic attribute binding, spatial relationships, motion binding, action binding, object interactions, and generative numeracy. We further carefully design evaluation metrics of multimodal large language model (MLLM)-based, detection-based, and tracking-based metrics, which can better reflect the compositional text-to-video generation quality of seven proposed categories with 1400 text prompts. The effectiveness of the proposed metrics is verified by correlation with human evaluations. We also benchmark various text-to-video generative models and conduct in-depth analysis across different models and various compositional categories. We find that compositional text-to-video generation is highly challenging for current models, and we hope our attempt could shed light on future research in this direction.
comment: Project page: https://t2v-compbench-2025.github.io/ Code: https://github.com/KaiyueSun98/T2V-CompBench/tree/V2
♻ ☆ DeblurDiNAT: A Compact Model with Exceptional Generalization and Visual Fidelity on Unseen Domains
Recent deblurring networks have effectively restored clear images from the blurred ones. However, they often struggle with generalization to unknown domains. Moreover, these models typically focus on distortion metrics such as PSNR and SSIM, neglecting the critical aspect of metrics aligned with human perception. To address these limitations, we propose DeblurDiNAT, a deblurring Transformer based on Dilated Neighborhood Attention. First, DeblurDiNAT employs an alternating dilation factor paradigm to capture both local and global blurred patterns, enhancing generalization and perceptual clarity. Second, a local cross-channel learner aids the Transformer block to understand the short-range relationships between adjacent channels. Additionally, we present a linear feed-forward network with a simple while effective design. Finally, a dual-stage feature fusion module is introduced as an alternative to the existing approach, which efficiently process multi-scale visual information across network levels. Compared to state-of-the-art models, our compact DeblurDiNAT demonstrates superior generalization capabilities and achieves remarkable performance in perceptual metrics, while maintaining a favorable model size.
♻ ☆ Click-Calib: A Robust Extrinsic Calibration Method for Surround-View Systems
Surround-View System (SVS) is an essential component in Advanced Driver Assistance System (ADAS) and requires precise calibrations. However, conventional offline extrinsic calibration methods are cumbersome and time-consuming as they rely heavily on physical patterns. Additionally, these methods primarily focus on short-range areas surrounding the vehicle, resulting in lower calibration quality in more distant zones. To address these limitations, we propose Click-Calib, a pattern-free approach for offline SVS extrinsic calibration. Without requiring any special setup, the user only needs to click a few keypoints on the ground in natural scenes. Unlike other offline calibration approaches, Click-Calib optimizes camera poses over a wide range by minimizing reprojection distance errors of keypoints, thereby achieving accurate calibrations at both short and long distances. Furthermore, Click-Calib supports both single-frame and multiple-frame modes, with the latter offering even better results. Evaluations on our in-house dataset and the public WoodScape dataset demonstrate its superior accuracy and robustness compared to baseline methods. Code is available at https://github.com/lwangvaleo/click_calib.
♻ ☆ A General Framework for Inference-time Scaling and Steering of Diffusion Models
Diffusion models produce impressive results in modalities ranging from images and video to protein design and text. However, generating samples with user-specified properties remains a challenge. Recent research proposes fine-tuning models to maximize rewards that capture desired properties, but these methods require expensive training and are prone to mode collapse. In this work, we propose Feynman Kac (FK) steering, an inference-time framework for steering diffusion models with reward functions. FK steering works by sampling a system of multiple interacting diffusion processes, called particles, and resampling particles at intermediate steps based on scores computed using functions called potentials. Potentials are defined using rewards for intermediate states and are selected such that a high value indicates that the particle will yield a high-reward sample. We explore various choices of potentials, intermediate rewards, and samplers. We evaluate FK steering on text-to-image and text diffusion models. For steering text-to-image models with a human preference reward, we find that FK steering a 0.8B parameter model outperforms a 2.6B parameter fine-tuned model on prompt fidelity, with faster sampling and no training. For steering text diffusion models with rewards for text quality and specific text attributes, we find that FK steering generates lower perplexity, more linguistically acceptable outputs and enables gradient-free control of attributes like toxicity. Our results demonstrate that inference-time scaling and steering of diffusion models, even with off-the-shelf rewards, can provide significant sample quality gains and controllability benefits. Code is available at https://github.com/zacharyhorvitz/Fk-Diffusion-Steering .
♻ ☆ SA-MLP: A Low-Power Multiplication-Free Deep Network for 3D Point Cloud Classification in Resource-Constrained Environments
Point cloud classification plays a crucial role in the processing and analysis of data from 3D sensors such as LiDAR, which are commonly used in applications like autonomous vehicles, robotics, and environmental monitoring. However, traditional neural networks, which rely heavily on multiplication operations, often face challenges in terms of high computational costs and energy consumption. This study presents a novel family of efficient MLP-based architectures designed to improve the computational efficiency of point cloud classification tasks in sensor systems. The baseline model, Mul-MLP, utilizes conventional multiplication operations, while Add-MLP and Shift-MLP replace multiplications with addition and shift operations, respectively. These replacements leverage more sensor-friendly operations that can significantly reduce computational overhead, making them particularly suitable for resource-constrained sensor platforms. To further enhance performance, we propose SA-MLP, a hybrid architecture that alternates between shift and adder layers, preserving the network depth while optimizing computational efficiency. Unlike previous approaches such as ShiftAddNet, which increase the layer count and limit representational capacity by freezing shift weights, SA-MLP fully exploits the complementary advantages of shift and adder layers by employing distinct learning rates and optimizers. Experimental results show that Add-MLP and Shift-MLP achieve competitive performance compared to Mul-MLP, while SA-MLP surpasses the baseline, delivering results comparable to state-of-the-art MLP models in terms of both classification accuracy and computational efficiency. This work offers a promising, energy-efficient solution for sensor-driven applications requiring real-time point cloud classification, particularly in environments with limited computational resources.
♻ ☆ A design of Convolutional Neural Network model for the Diagnosis of the COVID-19
With the spread of COVID-19 around the globe over the past year, the usage of artificial intelligence (AI) algorithms and image processing methods to analyze the X-ray images of patients' chest with COVID-19 has become essential. The COVID-19 virus recognition in the lung area of a patient is one of the basic and essential needs of clicical centers and hospitals. Most research in this field has been devoted to papers on the basis of deep learning methods utilizing CNNs (Convolutional Neural Network), which mainly deal with the screening of sick and healthy people.In this study, a new structure of a 19-layer CNN has been recommended for accurately recognition of the COVID-19 from the X-ray pictures of chest. The offered CNN is developed to serve as a precise diagnosis system for a three class (viral pneumonia, Normal, COVID) and a four classclassification (Lung opacity, Normal, COVID-19, and pneumonia). A comparison is conducted among the outcomes of the offered procedure and some popular pretrained networks, including Inception, Alexnet, ResNet50, Squeezenet, and VGG19 and based on Specificity, Accuracy, Precision, Sensitivity, Confusion Matrix, and F1-score. The experimental results of the offered CNN method specify its dominance over the existing published procedures. This method can be a useful tool for clinicians in deciding properly about COVID-19.
comment: Important mistakes found. There's no new version currently. Also contradiction with authorship
♻ ☆ Compression with Global Guidance: Towards Training-free High-Resolution MLLMs Acceleration
Multimodal large language models (MLLMs) have attracted considerable attention due to their exceptional performance in visual content understanding and reasoning. However, their inference efficiency has been a notable concern, as the increasing length of multimodal contexts leads to quadratic complexity. Token compression techniques, which reduce the number of visual tokens, have demonstrated their effectiveness in reducing computational costs. Yet, these approaches have struggled to keep pace with the rapid advancements in MLLMs, especially the AnyRes strategy in the context of high-resolution image understanding. In this paper, we propose a novel token compression method, GlobalCom$^2$, tailored for high-resolution MLLMs that receive both the thumbnail and multiple crops. GlobalCom$^2$ treats the tokens derived from the thumbnail as the "commander" of the entire token compression process, directing the allocation of retention ratios and the specific compression for each crop. In this way, redundant tokens are eliminated while important local details are adaptively preserved to the highest extent feasible. Empirical results across 10 benchmarks reveal that GlobalCom$^2$ achieves an optimal balance between performance and efficiency, and consistently outperforms state-of-the-art token compression methods with LLaVA-NeXT-7B/13B models. Our code is released at https://github.com/xuyang-liu16/GlobalCom2.
comment: Our code is released at \url{https://github.com/xuyang-liu16/GlobalCom2}
♻ ☆ Identifying Spurious Correlations using Counterfactual Alignment
Models driven by spurious correlations often yield poor generalization performance. We propose the counterfactual (CF) alignment method to detect and quantify spurious correlations of black box classifiers. Our methodology is based on counterfactual images generated with respect to one classifier being input into other classifiers to see if they also induce changes in the outputs of these classifiers. The relationship between these responses can be quantified and used to identify specific instances where a spurious correlation exists. This is validated by observing intuitive trends in face-attribute and waterbird classifiers, as well as by fabricating spurious correlations and detecting their presence, both visually and quantitatively. Furthermore, utilizing the CF alignment method, we demonstrate that we can evaluate robust optimization methods (GroupDRO, JTT, and FLAC) by detecting a reduction in spurious correlations.
comment: Accepted to Transactions on Machine Learning Research (TMLR), Code: https://github.com/ieee8023/latentshift
♻ ☆ PACE: Marrying generalization in PArameter-efficient fine-tuning with Consistency rEgularization NeurIPS 2024
Parameter-Efficient Fine-Tuning (PEFT) effectively adapts pre-trained transformers to downstream tasks. However, the optimization of tasks performance often comes at the cost of generalizability in fine-tuned models. To address this issue, we theoretically connect smaller weight gradient norms during training and larger datasets to the improvements in model generalization. Motivated by this connection, we propose reducing gradient norms for enhanced generalization and aligning fine-tuned model with the pre-trained counterpart to retain knowledge from large-scale pre-training data. Yet, naive alignment does not guarantee gradient reduction and can potentially cause gradient explosion, complicating efforts to manage gradients. To address such an issue, we propose PACE, marrying generalization of PArameter-efficient fine-tuning with Consistency rEgularization. We perturb features learned from the adapter with the multiplicative noise and ensure the fine-tuned model remains consistent for same sample under different perturbations. Theoretical analysis shows that PACE not only implicitly regularizes gradients for enhanced generalization, but also implicitly aligns the fine-tuned and pre-trained models to retain knowledge. Experimental evidence supports our theories. PACE surpasses existing PEFT methods in visual adaptation tasks (VTAB-1k, FGVC, few-shot learning, domain adaptation) showcasing its potential for resource-efficient fine-tuning. It also improves LoRA in text classification (GLUE) and mathematical reasoning (GSM-8K). The code is available at https://github.com/MaxwellYaoNi/PACE
comment: Accepted by NeurIPS 2024 as a spotlight
♻ ☆ TextSleuth: Towards Explainable Tampered Text Detection
Recently, tampered text detection has attracted increasing attention due to its essential role in information security. Although existing methods can detect the tampered text region, the interpretation of such detection remains unclear, making the prediction unreliable. To address this problem, we propose to explain the basis of tampered text detection with natural language via large multimodal models. To fill the data gap for this task, we propose a large-scale, comprehensive dataset, ETTD, which contains both pixel-level annotations for tampered text region and natural language annotations describing the anomaly of the tampered text. Multiple methods are employed to improve the quality of the proposed data. For example, elaborate queries are introduced to generate high-quality anomaly descriptions with GPT4o. A fused mask prompt is proposed to reduce confusion when querying GPT4o to generate anomaly descriptions. To automatically filter out low-quality annotations, we also propose to prompt GPT4o to recognize tampered texts before describing the anomaly, and to filter out the responses with low OCR accuracy. To further improve explainable tampered text detection, we propose a simple yet effective model called TextSleuth, which achieves improved fine-grained perception and cross-domain generalization by focusing on the suspected region, with a two-stage analysis paradigm and an auxiliary grounding prompt. Extensive experiments on both the ETTD dataset and the public dataset have verified the effectiveness of the proposed methods. In-depth analysis is also provided to inspire further research. Our dataset and code will be open-source.
comment: The first work for explainable tampered text detection
♻ ☆ A Foundation Language-Image Model of the Retina (FLAIR): Encoding Expert Knowledge in Text Supervision
Foundation vision-language models are currently transforming computer vision, and are on the rise in medical imaging fueled by their very promising generalization capabilities. However, the initial attempts to transfer this new paradigm to medical imaging have shown less impressive performances than those observed in other domains, due to the significant domain shift and the complex, expert domain knowledge inherent to medical-imaging tasks. Motivated by the need for domain-expert foundation models, we present FLAIR, a pre-trained vision-language model for universal retinal fundus image understanding. To this end, we compiled 38 open-access, mostly categorical fundus imaging datasets from various sources, with up to 101 different target conditions and 288,307 images. We integrate the expert's domain knowledge in the form of descriptive textual prompts, during both pre-training and zero-shot inference, enhancing the less-informative categorical supervision of the data. Such a textual expert's knowledge, which we compiled from the relevant clinical literature and community standards, describes the fine-grained features of the pathologies as well as the hierarchies and dependencies between them. We report comprehensive evaluations, which illustrate the benefit of integrating expert knowledge and the strong generalization capabilities of FLAIR under difficult scenarios with domain shifts or unseen categories. When adapted with a lightweight linear probe, FLAIR outperforms fully-trained, dataset-focused models, more so in the few-shot regimes. Interestingly, FLAIR outperforms by a wide margin larger-scale generalist image-language models and retina domain-specific self-supervised networks, which emphasizes the potential of embedding experts' domain knowledge and the limitations of generalist models in medical imaging.
comment: Accepted in Medical Image Analysis. The pre-trained model is available at: https://github.com/jusiro/FLAIR
♻ ☆ MADiff: Text-Guided Fashion Image Editing with Mask Prediction and Attention-Enhanced Diffusion
Text-guided image editing model has achieved great success in general domain. However, directly applying these models to the fashion domain may encounter two issues: (1) Inaccurate localization of editing region; (2) Weak editing magnitude. To address these issues, the MADiff model is proposed. Specifically, to more accurately identify editing region, the MaskNet is proposed, in which the foreground region, densepose and mask prompts from large language model are fed into a lightweight UNet to predict the mask for editing region. To strengthen the editing magnitude, the Attention-Enhanced Diffusion Model is proposed, where the noise map, attention map, and the mask from MaskNet are fed into the proposed Attention Processor to produce a refined noise map. By integrating the refined noise map into the diffusion model, the edited image can better align with the target prompt. Given the absence of benchmarks in fashion image editing, we constructed a dataset named Fashion-E, comprising 28390 image-text pairs in the training set, and 2639 image-text pairs for four types of fashion tasks in the evaluation set. Extensive experiments on Fashion-E demonstrate that our proposed method can accurately predict the mask of editing region and significantly enhance editing magnitude in fashion image editing compared to the state-of-the-art methods.
♻ ☆ Industrial Anomaly Detection and Localization Using Weakly-Supervised Residual Transformers
Recent advancements in industrial anomaly detection (AD) have demonstrated that incorporating a small number of anomalous samples during training can significantly enhance accuracy. However, this improvement often comes at the cost of extensive annotation efforts, which are impractical for many real-world applications. In this paper, we introduce a novel framework, Weak}ly-supervised RESidual Transformer (WeakREST), designed to achieve high anomaly detection accuracy while minimizing the reliance on manual annotations. First, we reformulate the pixel-wise anomaly localization task into a block-wise classification problem. Second, we introduce a residual-based feature representation called Positional Fast Anomaly Residuals (PosFAR) which captures anomalous patterns more effectively. To leverage this feature, we adapt the Swin Transformer for enhanced anomaly detection and localization. Additionally, we propose a weak annotation approach, utilizing bounding boxes and image tags to define anomalous regions. This approach establishes a semi-supervised learning context that reduces the dependency on precise pixel-level labels. To further improve the learning process, we develop a novel ResMixMatch algorithm, capable of handling the interplay between weak labels and residual-based representations. On the benchmark dataset MVTec-AD, our method achieves an Average Precision (AP) of $83.0\%$, surpassing the previous best result of $82.7\%$ in the unsupervised setting. In the supervised AD setting, WeakREST attains an AP of $87.6\%$, outperforming the previous best of $86.0\%$. Notably, even when using weaker annotations such as bounding boxes, WeakREST exceeds the performance of leading methods relying on pixel-wise supervision, achieving an AP of $87.1\%$ compared to the prior best of $86.0\%$ on MVTec-AD.
comment: 13 pages,7 figures
♻ ☆ The Surprising Ineffectiveness of Pre-Trained Visual Representations for Model-Based Reinforcement Learning NeurIPS 2024
Visual Reinforcement Learning (RL) methods often require extensive amounts of data. As opposed to model-free RL, model-based RL (MBRL) offers a potential solution with efficient data utilization through planning. Additionally, RL lacks generalization capabilities for real-world tasks. Prior work has shown that incorporating pre-trained visual representations (PVRs) enhances sample efficiency and generalization. While PVRs have been extensively studied in the context of model-free RL, their potential in MBRL remains largely unexplored. In this paper, we benchmark a set of PVRs on challenging control tasks in a model-based RL setting. We investigate the data efficiency, generalization capabilities, and the impact of different properties of PVRs on the performance of model-based agents. Our results, perhaps surprisingly, reveal that for MBRL current PVRs are not more sample efficient than learning representations from scratch, and that they do not generalize better to out-of-distribution (OOD) settings. To explain this, we analyze the quality of the trained dynamics model. Furthermore, we show that data diversity and network architecture are the most important contributors to OOD generalization performance.
comment: Published at the 38th Conference on Neural Information Processing Systems (NeurIPS 2024). Project page: https://schneimo.com/pvr4mbrl/
♻ ☆ CGCOD: Class-Guided Camouflaged Object Detection
Camouflaged Object Detection (COD) aims to identify objects that blend seamlessly into their surroundings. The inherent visual complexity of camouflaged objects, including their low contrast with the background, diverse textures, and subtle appearance variations, often obscures semantic cues, making accurate segmentation highly challenging. Existing methods primarily rely on visual features, which are insufficient to handle the variability and intricacy of camouflaged objects, leading to unstable object perception and ambiguous segmentation results. To tackle these limitations, we introduce a novel task, class-guided camouflaged object detection (CGCOD), which extends traditional COD task by incorporating object-specific class knowledge to enhance detection robustness and accuracy. To facilitate this task, we present a new dataset, CamoClass, comprising real-world camouflaged objects with class annotations. Furthermore, we propose a multi-stage framework, CGNet, which incorporates a plug-and-play class prompt generator and a simple yet effective class-guided detector. This establishes a new paradigm for COD, bridging the gap between contextual understanding and class-guided detection. Extensive experimental results demonstrate the effectiveness of our flexible framework in improving the performance of proposed and existing detectors by leveraging class-level textual information.
♻ ☆ Evaluation of radiomic feature harmonization techniques for benign and malignant pulmonary nodules
BACKGROUND: Radiomics provides quantitative features of pulmonary nodules (PNs) which could aid lung cancer diagnosis, but medical image acquisition variability is an obstacle to clinical application. Acquisition effects may differ between radiomic features from benign vs. malignant PNs. PURPOSE: We evaluated how to account for differences between benign and malignant PNs when correcting radiomic features' acquisition dependency. METHODS: We used 567 chest CT scans grouped as benign, malignant, or lung cancer screening (mixed benign, malignant). ComBat harmonization was applied to extracted features for variation in 4 acquisition parameters. We compared: harmonizing without distinction, harmonizing with a covariate to preserve distinctions between subgroups, and harmonizing subgroups separately. Significant ($p\le0.05$) Kruskal-Wallis tests showed whether harmonization removed acquisition dependency. A LASSO-SVM pipeline was trained on successfully harmonized features to predict malignancy. To evaluate predictive information in these features, the trained harmonization estimators and predictive model were applied to unseen test sets. Harmonization and predictive performance were assessed for 10 trials of 5-fold cross-validation. RESULTS: An average 2.1% of features (95% CI:1.9-2.4%) were acquisition-independent when harmonized without distinction, 27.3% (95% CI:25.7-28.9%) when harmonized with a covariate, and 90.9% (95% CI:90.4-91.5%) when harmonized separately. Data harmonized separately or with a covariate trained models with higher ROC-AUC for screening scans than data harmonized without distinction between benign and malignant PNs (Delong test, adjusted $p\le0.05$). CONCLUSIONS: Radiomic features of benign and malignant PNs need different corrective transformations to recover acquisition-independent distributions. This can be done by harmonizing separately or with a covariate.
comment: 15 pages, 3 figures, plus supplemental material; updated author list, corrected result in paragraph 3 of Discussion, updated Figure S1
♻ ☆ Structural damage detection via hierarchical damage information with volumetric assessment
Structural health monitoring (SHM) is essential for ensuring the safety and longevity of infrastructure, but complex image environments, noisy labels, and reliance on manual damage assessments often hinder its effectiveness. This study introduces the Guided Detection Network (Guided-DetNet), a framework designed to address these challenges. Guided-DetNet is characterized by a Generative Attention Module (GAM), Hierarchical Elimination Algorithm (HEA), and Volumetric Contour Visual Assessment (VCVA). GAM leverages cross-horizontal and cross-vertical patch merging and cross-foreground-background feature fusion to generate varied features to mitigate complex image environments. HEA addresses noisy labeling using hierarchical relationships among classes to refine instances given an image by eliminating unlikely class instances. VCVA assesses the severity of detected damages via volumetric representation and quantification leveraging the Dirac delta distribution. A comprehensive quantitative study and two robustness tests were conducted using the PEER Hub dataset, and a drone-based application, which involved a field experiment, was conducted to substantiate Guided-DetNet's promising performances. In triple classification tasks, the framework achieved 96% accuracy, surpassing state-of-the-art classifiers by up to 3%. In dual detection tasks, it outperformed competitive detectors with a precision of 94% and a mean average precision (mAP) of 79% while maintaining a frame rate of 57.04fps, suitable for real-time applications. Additionally, robustness tests demonstrated resilience under adverse conditions, with precision scores ranging from 79% to 91%. Guided-DetNet is established as a robust and efficient framework for SHM, offering advancements in automation and precision, with the potential for widespread application in drone-based infrastructure inspections.
♻ ☆ SemTalk: Holistic Co-speech Motion Generation with Frame-level Semantic Emphasis
A good co-speech motion generation cannot be achieved without a careful integration of common rhythmic motion and rare yet essential semantic motion. In this work, we propose SemTalk for holistic co-speech motion generation with frame-level semantic emphasis. Our key insight is to separately learn general motions and sparse motions, and then adaptively fuse them. In particular, rhythmic consistency learning is explored to establish rhythm-related base motion, ensuring a coherent foundation that synchronizes gestures with the speech rhythm. Subsequently, textit{semantic emphasis learning is designed to generate semantic-aware sparse motion, focusing on frame-level semantic cues. Finally, to integrate sparse motion into the base motion and generate semantic-emphasized co-speech gestures, we further leverage a learned semantic score for adaptive synthesis. Qualitative and quantitative comparisons on two public datasets demonstrate that our method outperforms the state-of-the-art, delivering high-quality co-speech motion with enhanced semantic richness over a stable base motion.
comment: 11 pages, 8 figures
♻ ☆ ACE++: Instruction-Based Image Creation and Editing via Context-Aware Content Filling
We report ACE++, an instruction-based diffusion framework that tackles various image generation and editing tasks. Inspired by the input format for the inpainting task proposed by FLUX.1-Fill-dev, we improve the Long-context Condition Unit (LCU) introduced in ACE and extend this input paradigm to any editing and generation tasks. To take full advantage of image generative priors, we develop a two-stage training scheme to minimize the efforts of finetuning powerful text-to-image diffusion models like FLUX.1-dev. In the first stage, we pre-train the model using task data with the 0-ref tasks from the text-to-image model. There are many models in the community based on the post-training of text-to-image foundational models that meet this training paradigm of the first stage. For example, FLUX.1-Fill-dev deals primarily with painting tasks and can be used as an initialization to accelerate the training process. In the second stage, we finetune the above model to support the general instructions using all tasks defined in ACE. To promote the widespread application of ACE++ in different scenarios, we provide a comprehensive set of models that cover both full finetuning and lightweight finetuning, while considering general applicability and applicability in vertical scenarios. The qualitative analysis showcases the superiority of ACE++ in terms of generating image quality and prompt following ability. Code and models will be available on the project page: https://ali-vilab. github.io/ACE_plus_page/.
♻ ☆ Solving Energy-Independent Density for CT Metal Artifact Reduction via Neural Representation
X-ray CT often suffers from shadowing and streaking artifacts in the presence of metallic materials, which severely degrade imaging quality. Physically, the linear attenuation coefficients (LACs) of metals vary significantly with X-ray energy, causing a nonlinear beam hardening effect (BHE) in CT measurements. Reconstructing CT images from metal-corrupted measurements consequently becomes a challenging nonlinear inverse problem. Existing state-of-the-art (SOTA) metal artifact reduction (MAR) algorithms rely on supervised learning with numerous paired CT samples. While promising, these supervised methods often assume that the unknown LACs are energy-independent, ignoring the energy-induced BHE, which results in limited generalization. Moreover, the requirement for large datasets also limits their applications in real-world scenarios. In this work, we propose Density neural representation (Diner), a novel unsupervised MAR method. Our key innovation lies in formulating MAR as an energy-independent density reconstruction problem that strictly adheres to the photon-tissue absorption physical model. This model is inherently nonlinear and complex, making it a rarely considered approach in inverse imaging problems. By introducing the water-equivalent tissues approximation and a new polychromatic model to characterize the nonlinear CT acquisition process, we directly learn the neural representation of the density map from raw measurements without using external training data. This energy-independent density reconstruction framework fundamentally resolves the nonlinear BHE, enabling superior MAR performance across a wide range of scanning scenarios. Extensive experiments on both simulated and real-world datasets demonstrate the superiority of our unsupervised Diner over popular supervised methods in terms of MAR performance and robustness.
comment: 11 pages
♻ ☆ 3VL: Using Trees to Improve Vision-Language Models' Interpretability
Vision-Language models (VLMs) have proven to be effective at aligning image and text representations, producing superior zero-shot results when transferred to many downstream tasks. However, these representations suffer from some key shortcomings in understanding Compositional Language Concepts (CLC), such as recognizing objects' attributes, states, and relations between different objects. Moreover, VLMs typically have poor interpretability, making it challenging to debug and mitigate compositional-understanding failures. In this work, we introduce the architecture and training technique of Tree-augmented Vision-Language (3VL) model accompanied by our proposed Anchor inference method and Differential Relevance (DiRe) interpretability tool. By expanding the text of an arbitrary image-text pair into a hierarchical tree structure using language analysis tools, 3VL allows the induction of this structure into the visual representation learned by the model, enhancing its interpretability and compositional reasoning. Additionally, we show how Anchor, a simple technique for text unification, can be used to filter nuisance factors while increasing CLC understanding performance, e.g., on the fundamental VL-Checklist benchmark. We also show how DiRe, which performs a differential comparison between VLM relevancy maps, enables us to generate compelling visualizations of the reasons for a model's success or failure. Our code is available at: https://github.com/niryellinek/3VL.
comment: accepted to IEEE TIP
♻ ☆ When No-Reference Image Quality Models Meet MAP Estimation in Diffusion Latents
Contemporary no-reference image quality assessment (NR-IQA) models can effectively quantify perceived image quality, often achieving strong correlations with human perceptual scores on standard IQA benchmarks. Yet, limited efforts have been devoted to treating NR-IQA models as natural image priors for real-world image enhancement, and consequently comparing them from a perceptual optimization standpoint. In this work, we show -- for the first time -- that NR-IQA models can be plugged into the maximum a posteriori (MAP) estimation framework for image enhancement. This is achieved by performing gradient ascent in the diffusion latent space rather than in the raw pixel domain, leveraging a pretrained differentiable and bijective diffusion process. Likely, different NR-IQA models lead to different enhanced outputs, which in turn provides a new computational means of comparing them. Unlike conventional correlation-based measures, our comparison method offers complementary insights into the respective strengths and weaknesses of the competing NR-IQA models in perceptual optimization scenarios. Additionally, we aim to improve the best-performing NR-IQA model in diffusion latent MAP estimation by incorporating the advantages of other top-performing methods. The resulting model delivers noticeably better results in enhancing real-world images afflicted by unknown and complex distortions, all preserving a high degree of image fidelity.
♻ ☆ Sports-QA: A Large-Scale Video Question Answering Benchmark for Complex and Professional Sports
Reasoning over sports videos for question answering is an important task with numerous applications, such as player training and information retrieval. However, this task has not been explored due to the lack of relevant datasets and the challenging nature it presents. Most datasets for video question answering (VideoQA) focus mainly on general and coarse-grained understanding of daily-life videos, which is not applicable to sports scenarios requiring professional action understanding and fine-grained motion analysis. In this paper, we introduce the first dataset, named Sports-QA, specifically designed for the sports VideoQA task. The Sports-QA dataset includes various types of questions, such as descriptions, chronologies, causalities, and counterfactual conditions, covering multiple sports. Furthermore, to address the characteristics of the sports VideoQA task, we propose a new Auto-Focus Transformer (AFT) capable of automatically focusing on particular scales of temporal information for question answering. We conduct extensive experiments on Sports-QA, including baseline studies and the evaluation of different methods. The results demonstrate that our AFT achieves state-of-the-art performance.
♻ ☆ Maximizing Uncertainty for Federated learning via Bayesian Optimisation-based Model Poisoning
As we transition from Narrow Artificial Intelligence towards Artificial Super Intelligence, users are increasingly concerned about their privacy and the trustworthiness of machine learning (ML) technology. A common denominator for the metrics of trustworthiness is the quantification of uncertainty inherent in DL algorithms, and specifically in the model parameters, input data, and model predictions. One of the common approaches to address privacy-related issues in DL is to adopt distributed learning such as federated learning (FL), where private raw data is not shared among users. Despite the privacy-preserving mechanisms in FL, it still faces challenges in trustworthiness. Specifically, the malicious users, during training, can systematically create malicious model parameters to compromise the models predictive and generative capabilities, resulting in high uncertainty about their reliability. To demonstrate malicious behaviour, we propose a novel model poisoning attack method named Delphi which aims to maximise the uncertainty of the global model output. We achieve this by taking advantage of the relationship between the uncertainty and the model parameters of the first hidden layer of the local model. Delphi employs two types of optimisation , Bayesian Optimisation and Least Squares Trust Region, to search for the optimal poisoned model parameters, named as Delphi-BO and Delphi-LSTR. We quantify the uncertainty using the KL Divergence to minimise the distance of the predictive probability distribution towards an uncertain distribution of model output. Furthermore, we establish a mathematical proof for the attack effectiveness demonstrated in FL. Numerical results demonstrate that Delphi-BO induces a higher amount of uncertainty than Delphi-LSTR highlighting vulnerability of FL systems to model poisoning attacks.
comment: 14 pages
♻ ☆ MGF: Mixed Gaussian Flow for Diverse Trajectory Prediction
To predict future trajectories, the normalizing flow with a standard Gaussian prior suffers from weak diversity. The ineffectiveness comes from the conflict between the fact of asymmetric and multi-modal distribution of likely outcomes and symmetric and single-modal original distribution and supervision losses. Instead, we propose constructing a mixed Gaussian prior for a normalizing flow model for trajectory prediction. The prior is constructed by analyzing the trajectory patterns in the training samples without requiring extra annotations while showing better expressiveness and being multi-modal and asymmetric. Besides diversity, it also provides better controllability for probabilistic trajectory generation. We name our method Mixed Gaussian Flow (MGF). It achieves state-of-the-art performance in the evaluation of both trajectory alignment and diversity on the popular UCY/ETH and SDD datasets. Code is available at https://github.com/mulplue/MGF.
comment: Accepted by Neurips 2024. Code: https://github.com/mulplue/MGF
♻ ☆ Mask-guided cross-image attention for zero-shot in-silico histopathologic image generation with a diffusion model
Creating in-silico data with generative AI promises a cost-effective alternative to staining, imaging, and annotating whole slide images in computational pathology. Diffusion models are the state-of-the-art solution for generating in-silico images, offering unparalleled fidelity and realism. Using appearance transfer diffusion models allows for zero-shot image generation, facilitating fast application and making model training unnecessary. However current appearance transfer diffusion models are designed for natural images, where the main task is to transfer the foreground object from an origin to a target domain, while the background is of insignificant importance. In computational pathology, specifically in oncology, it is however not straightforward to define which objects in an image should be classified as foreground and background, as all objects in an image may be of critical importance for the detailed understanding the tumor micro-environment. We contribute to the applicability of appearance transfer diffusion models to immunohistochemistry-stained images by modifying the appearance transfer guidance to alternate between class-specific AdaIN feature statistics matchings using existing segmentation masks. The performance of the proposed method is demonstrated on the downstream task of supervised epithelium segmentation, showing that the number of manual annotations required for model training can be reduced by 75%, outperforming the baseline approach. Additionally, we consulted with a certified pathologist to investigate future improvements. We anticipate this work to inspire the application of zero-shot diffusion models in computational pathology, providing an efficient method to generate in-silico images with unmatched fidelity and realism, which prove meaningful for downstream tasks, such as training existing deep learning models or finetuning foundation models.
comment: 5 pages
♻ ☆ RoHan: Robust Hand Detection in Operation Room
Hand-specific localization has garnered significant interest within the computer vision community. Although there are numerous datasets with hand annotations from various angles and settings, domain transfer techniques frequently struggle in surgical environments. This is mainly due to the limited availability of gloved hand instances and the unique challenges of operating rooms (ORs). Thus, hand-detection models tailored to OR settings require extensive training and expensive annotation processes. To overcome these challenges, we present "RoHan" - a novel approach for robust hand detection in the OR, leveraging advanced semi-supervised domain adaptation techniques to tackle the challenges of varying recording conditions, diverse glove colors, and occlusions common in surgical settings. Our methodology encompasses two main stages: (1) data augmentation strategy that utilizes "Artificial Gloves," a method for augmenting publicly available hand datasets with synthetic images of hands-wearing gloves; (2) semi-supervised domain adaptation pipeline that improves detection performance in real-world OR settings through iterative prediction refinement and efficient frame filtering. We evaluate our method using two datasets: simulated enterotomy repair and saphenous vein graft harvesting. "RoHan" substantially reduces the need for extensive labeling and model training, paving the way for the practical implementation of hand detection technologies in medical settings.
comment: 12 pages
♻ ☆ Diffusion-based Unsupervised Audio-visual Speech Enhancement
This paper proposes a new unsupervised audio-visual speech enhancement (AVSE) approach that combines a diffusion-based audio-visual speech generative model with a non-negative matrix factorization (NMF) noise model. First, the diffusion model is pre-trained on clean speech conditioned on corresponding video data to simulate the speech generative distribution. This pre-trained model is then paired with the NMF-based noise model to estimate clean speech iteratively. Specifically, a diffusion-based posterior sampling approach is implemented within the reverse diffusion process, where after each iteration, a speech estimate is obtained and used to update the noise parameters. Experimental results confirm that the proposed AVSE approach not only outperforms its audio-only counterpart but also generalizes better than a recent supervised-generative AVSE method. Additionally, the new inference algorithm offers a better balance between inference speed and performance compared to the previous diffusion-based method. Code and demo available at: https://jeaneudesayilo.github.io/fast_UdiffSE
♻ ☆ Improving Pain Classification using Spatio-Temporal Deep Learning Approaches with Facial Expressions
Pain management and severity detection are crucial for effective treatment, yet traditional self-reporting methods are subjective and may be unsuitable for non-verbal individuals (people with limited speaking skills). To address this limitation, we explore automated pain detection using facial expressions. Our study leverages deep learning techniques to improve pain assessment by analyzing facial images from the Pain Emotion Faces Database (PEMF). We propose two novel approaches1: (1) a hybrid ConvNeXt model combined with Long Short-Term Memory (LSTM) blocks to analyze video frames and predict pain presence, and (2) a Spatio-Temporal Graph Convolution Network (STGCN) integrated with LSTM to process landmarks from facial images for pain detection. Our work represents the first use of the PEMF dataset for binary pain classification and demonstrates the effectiveness of these models through extensive experimentation. The results highlight the potential of combining spatial and temporal features for enhanced pain detection, offering a promising advancement in objective pain assessment methodologies.
comment: 8 pages, 3 figures, 3 tables. Accepted and presented at the 18th International Conference on Machine Vision (ICMV 2024), Edinburgh, UK
♻ ☆ Multispectral Pedestrian Detection with Sparsely Annotated Label AAAI 2025
Although existing Sparsely Annotated Object Detection (SAOD) approches have made progress in handling sparsely annotated environments in multispectral domain, where only some pedestrians are annotated, they still have the following limitations: (i) they lack considerations for improving the quality of pseudo-labels for missing annotations, and (ii) they rely on fixed ground truth annotations, which leads to learning only a limited range of pedestrian visual appearances in the multispectral domain. To address these issues, we propose a novel framework called Sparsely Annotated Multispectral Pedestrian Detection (SAMPD). For limitation (i), we introduce Multispectral Pedestrian-aware Adaptive Weight (MPAW) and Positive Pseudo-label Enhancement (PPE) module. Utilizing multispectral knowledge, these modules ensure the generation of high-quality pseudo-labels and enable effective learning by increasing weights for high-quality pseudo-labels based on modality characteristics. To address limitation (ii), we propose an Adaptive Pedestrian Retrieval Augmentation (APRA) module, which adaptively incorporates pedestrian patches from ground-truth and dynamically integrates high-quality pseudo-labels with the ground-truth, facilitating a more diverse learning pool of pedestrians. Extensive experimental results demonstrate that our SAMPD significantly enhances performance in sparsely annotated environments within the multispectral domain.
comment: Accepted at AAAI 2025
♻ ☆ Approximation properties relative to continuous scale space for hybrid discretizations of Gaussian derivative operators
This paper presents an analysis of properties of two hybrid discretization methods for Gaussian derivatives, based on convolutions with either the normalized sampled Gaussian kernel or the integrated Gaussian kernel followed by central differences. The motivation for studying these discretization methods is that in situations when multiple spatial derivatives of different order are needed at the same scale level, they can be computed significantly more efficiently compared to more direct derivative approximations based on explicit convolutions with either sampled Gaussian kernels or integrated Gaussian kernels. While these computational benefits do also hold for the genuinely discrete approach for computing discrete analogues of Gaussian derivatives, based on convolution with the discrete analogue of the Gaussian kernel followed by central differences, the underlying mathematical primitives for the discrete analogue of the Gaussian kernel, in terms of modified Bessel functions of integer order, may not be available in certain frameworks for image processing, such as when performing deep learning based on scale-parameterized filters in terms of Gaussian derivatives, with learning of the scale levels. In this paper, we present a characterization of the properties of these hybrid discretization methods, in terms of quantitative performance measures concerning the amount of spatial smoothing that they imply, as well as the relative consistency of scale estimates obtained from scale-invariant feature detectors with automatic scale selection, with an emphasis on the behaviour for very small values of the scale parameter, which may differ significantly from corresponding results obtained from the fully continuous scale-space theory, as well as between different types of discretization methods.
comment: 23 pages, 9 figures. arXiv admin note: text overlap with arXiv:2311.11317
♻ ☆ OminiControl: Minimal and Universal Control for Diffusion Transformer
In this paper, we introduce OminiControl, a highly versatile and parameter-efficient framework that integrates image conditions into pre-trained Diffusion Transformer (DiT) models. At its core, OminiControl leverages a parameter reuse mechanism, enabling the DiT to encode image conditions using itself as a powerful backbone and process them with its flexible multi-modal attention processors. Unlike existing methods, which rely heavily on additional encoder modules with complex architectures, OminiControl (1) effectively and efficiently incorporates injected image conditions with only ~0.1% additional parameters, and (2) addresses a wide range of image conditioning tasks in a unified manner, including subject-driven generation and spatially-aligned conditions such as edges, depth, and more. Remarkably, these capabilities are achieved by training on images generated by the DiT itself, which is particularly beneficial for subject-driven generation. Extensive evaluations demonstrate that OminiControl outperforms existing UNet-based and DiT-adapted models in both subject-driven and spatially-aligned conditional generation. Additionally, we release our training dataset, Subjects200K, a diverse collection of over 200,000 identity-consistent images, along with an efficient data synthesis pipeline to advance research in subject-consistent generation.
♻ ☆ CrossFi: A Cross Domain Wi-Fi Sensing Framework Based on Siamese Network
In recent years, Wi-Fi sensing has garnered significant attention due to its numerous benefits, such as privacy protection, low cost, and penetration ability. Extensive research has been conducted in this field, focusing on areas such as gesture recognition, people identification, and fall detection. However, many data-driven methods encounter challenges related to domain shift, where the model fails to perform well in environments different from the training data. One major factor contributing to this issue is the limited availability of Wi-Fi sensing datasets, which makes models learn excessive irrelevant information and over-fit to the training set. Unfortunately, collecting large-scale Wi-Fi sensing datasets across diverse scenarios is a challenging task. To address this problem, we propose CrossFi, a siamese network-based approach that excels in both in-domain scenario and cross-domain scenario, including few-shot, zero-shot scenarios, and even works in few-shot new-class scenario where testing set contains new categories. The core component of CrossFi is a sample-similarity calculation network called CSi-Net, which improves the structure of the siamese network by using an attention mechanism to capture similarity information, instead of simply calculating the distance or cosine similarity. Based on it, we develop an extra Weight-Net that can generate a template for each class, so that our CrossFi can work in different scenarios. Experimental results demonstrate that our CrossFi achieves state-of-the-art performance across various scenarios. In gesture recognition task, our CrossFi achieves an accuracy of 98.17% in in-domain scenario, 91.72% in one-shot cross-domain scenario, 64.81% in zero-shot cross-domain scenario, and 84.75% in one-shot new-class scenario. The code for our model is publicly available at https://github.com/RS2002/CrossFi.
♻ ☆ Multiple Information Prompt Learning for Cloth-Changing Person Re-Identification
Cloth-changing person re-identification is a subject closer to the real world, which focuses on solving the problem of person re-identification after pedestrians change clothes. The primary challenge in this field is to overcome the complex interplay between intra-class and inter-class variations and to identify features that remain unaffected by changes in appearance. Sufficient data collection for model training would significantly aid in addressing this problem. However, it is challenging to gather diverse datasets in practice. Current methods focus on implicitly learning identity information from the original image or introducing additional auxiliary models, which are largely limited by the quality of the image and the performance of the additional model. To address these issues, inspired by prompt learning, we propose a novel multiple information prompt learning (MIPL) scheme for cloth-changing person ReID, which learns identity robust features through the common prompt guidance of multiple messages. Specifically, the clothing information stripping (CIS) module is designed to decouple the clothing information from the original RGB image features to counteract the influence of clothing appearance. The Bio-guided attention (BGA) module is proposed to increase the learning intensity of the model for key information. A dual-length hybrid patch (DHP) module is employed to make the features have diverse coverage to minimize the impact of feature bias. Extensive experiments demonstrate that the proposed method outperforms all state-of-the-art methods on the LTCC, Celeb-reID, Celeb-reID-light, and CSCC datasets, achieving rank-1 scores of 74.8%, 73.3%, 66.0%, and 88.1%, respectively. When compared to AIM (CVPR23), ACID (TIP23), and SCNet (MM23), MIPL achieves rank-1 improvements of 11.3%, 13.8%, and 7.9%, respectively, on the PRCC dataset.
♻ ☆ The Silent Majority: Demystifying Memorization Effect in the Presence of Spurious Correlations
Machine learning models often rely on simple spurious features -- patterns in training data that correlate with targets but are not causally related to them, like image backgrounds in foreground classification. This reliance typically leads to imbalanced test performance across minority and majority groups. In this work, we take a closer look at the fundamental cause of such imbalanced performance through the lens of memorization, which refers to the ability to predict accurately on \textit{atypical} examples (minority groups) in the training set but failing in achieving the same accuracy in the testing set. This paper systematically shows the ubiquitous existence of spurious features in a small set of neurons within the network, providing the first-ever evidence that memorization may contribute to imbalanced group performance. Through three experimental sources of converging empirical evidence, we find the property of a small subset of neurons or channels in memorizing minority group information. Inspired by these findings, we articulate the hypothesis: the imbalanced group performance is a byproduct of ``noisy'' spurious memorization confined to a small set of neurons. To further substantiate this hypothesis, we show that eliminating these unnecessary spurious memorization patterns via a novel framework during training can significantly affect the model performance on minority groups. Our experimental results across various architectures and benchmarks offer new insights on how neural networks encode core and spurious knowledge, laying the groundwork for future research in demystifying robustness to spurious correlation.
♻ ☆ DATransNet: Dynamic Attention Transformer Network for Infrared Small Target Detection
Infrared small target detection (ISTD) is widely used in civilian and military applications. However, ISTD encounters several challenges, including the tendency for small and dim targets to be obscured by complex backgrounds.To address this issue, we propose the Dynamic Attention Transformer Network (DATransNet), which aims to extract and preserve edge information of small targets.DATransNet employs the Dynamic Attention Transformer (DATrans), simulating central difference convolutions (CDC) to extract and integrate gradient features with deeper features.Furthermore, we propose a global feature extraction module (GFEM) that offers a comprehensive perspective to prevent the network from focusing solely on details while neglecting the background information. We compare the network with state-of-the-art (SOTA) approaches, and the results demonstrate that our method performs effectively. Our source code is available at https://github.com/greekinRoma/DATransNet.
♻ ☆ Ultra-High-Definition Image Deblurring via Multi-scale Cubic-Mixer
Currently, transformer-based algorithms are making a splash in the domain of image deblurring. Their achievement depends on the self-attention mechanism with CNN stem to model long range dependencies between tokens. Unfortunately, this ear-pleasing pipeline introduces high computational complexity and makes it difficult to run an ultra-high-definition image on a single GPU in real time. To trade-off accuracy and efficiency, the input degraded image is computed cyclically over three dimensional ($C$, $W$, and $H$) signals without a self-attention mechanism. We term this deep network as Multi-scale Cubic-Mixer, which is acted on both the real and imaginary components after fast Fourier transform to estimate the Fourier coefficients and thus obtain a deblurred image. Furthermore, we combine the multi-scale cubic-mixer with a slicing strategy to generate high-quality results at a much lower computational cost. Experimental results demonstrate that the proposed algorithm performs favorably against the state-of-the-art deblurring approaches on the several benchmarks and a new ultra-high-definition dataset in terms of accuracy and speed.
comment: 9 pages
♻ ☆ Zero-shot Video Restoration and Enhancement Using Pre-Trained Image Diffusion Model AAAI 2025
Diffusion-based zero-shot image restoration and enhancement models have achieved great success in various tasks of image restoration and enhancement. However, directly applying them to video restoration and enhancement results in severe temporal flickering artifacts. In this paper, we propose the first framework for zero-shot video restoration and enhancement based on the pre-trained image diffusion model. By replacing the spatial self-attention layer with the proposed short-long-range (SLR) temporal attention layer, the pre-trained image diffusion model can take advantage of the temporal correlation between frames. We further propose temporal consistency guidance, spatial-temporal noise sharing, and an early stopping sampling strategy to improve temporally consistent sampling. Our method is a plug-and-play module that can be inserted into any diffusion-based image restoration or enhancement methods to further improve their performance. Experimental results demonstrate the superiority of our proposed method. Our code is available at https://github.com/cao-cong/ZVRD.
comment: Accepted by AAAI 2025
♻ ☆ Continuous Concepts Removal in Text-to-image Diffusion Models
Text-to-image diffusion models have shown an impressive ability to generate high-quality images from input textual descriptions. However, concerns have been raised about the potential for these models to create content that infringes on copyrights or depicts disturbing subject matter. Removing specific concepts from these models is a promising potential solution to this problem. However, existing methods for concept removal do not work well in practical but challenging scenarios where concepts need to be continuously removed. Specifically, these methods lead to poor alignment between the text prompts and the generated image after the continuous removal process. To address this issue, we propose a novel approach called CCRT that includes a designed knowledge distillation paradigm. It constrains the text-image alignment behavior during the continuous concept removal process by using a set of text prompts generated through our genetic algorithm, which employs a designed fuzzing strategy. We conduct extensive experiments involving the removal of various concepts. The results evaluated through both algorithmic metrics and human studies demonstrate that our CCRT can effectively remove the targeted concepts in a continuous manner while maintaining the high generation quality (e.g., text-image alignment) of the model.
♻ ☆ Conformal-in-the-Loop for Learning with Imbalanced Noisy Data
Class imbalance and label noise are pervasive in large-scale datasets, yet much of machine learning research assumes well-labeled, balanced data, which rarely reflects real world conditions. Existing approaches typically address either label noise or class imbalance in isolation, leading to suboptimal results when both issues coexist. In this work, we propose Conformal-in-the-Loop (CitL), a novel training framework that addresses both challenges with a conformal prediction-based approach. CitL evaluates sample uncertainty to adjust weights and prune unreliable examples, enhancing model resilience and accuracy with minimal computational cost. Our extensive experiments include a detailed analysis showing how CitL effectively emphasizes impactful data in noisy, imbalanced datasets. Our results show that CitL consistently boosts model performance, achieving up to a 6.1% increase in classification accuracy and a 5.0 mIoU improvement in segmentation. Our code is publicly available: CitL.
comment: Under Review
♻ ☆ Investigating the Effect of Network Pruning on Performance and Interpretability
Deep Neural Networks (DNNs) are often over-parameterized for their tasks and can be compressed quite drastically by removing weights, a process called pruning. We investigate the impact of different pruning techniques on the classification performance and interpretability of GoogLeNet. We systematically apply unstructured and structured pruning, as well as connection sparsity (pruning of input weights) methods to the network and analyze the outcomes regarding the network's performance on the validation set of ImageNet. We also compare different retraining strategies, such as iterative pruning and one-shot pruning. We find that with sufficient retraining epochs, the performance of the networks can approximate the performance of the default GoogLeNet - and even surpass it in some cases. To assess interpretability, we employ the Mechanistic Interpretability Score (MIS) developed by Zimmermann et al. . Our experiments reveal that there is no significant relationship between interpretability and pruning rate when using MIS as a measure. Additionally, we observe that networks with extremely low accuracy can still achieve high MIS scores, suggesting that the MIS may not always align with intuitive notions of interpretability, such as understanding the basis of correct decisions.
comment: 4 pages, 6 figures
♻ ☆ Make-A-Character 2: Animatable 3D Character Generation From a Single Image
This report introduces Make-A-Character 2, an advanced system for generating high-quality 3D characters from single portrait photographs, ideal for game development and digital human applications. Make-A-Character 2 builds upon its predecessor by incorporating several significant improvements for image-based head generation. We utilize the IC-Light method to correct non-ideal illumination in input photos and apply neural network-based color correction to harmonize skin tones between the photos and game engine renders. We also employ the Hierarchical Representation Network to capture high-frequency facial structures and conduct adaptive skeleton calibration for accurate and expressive facial animations. The entire image-to-3D-character generation process takes less than 2 minutes. Furthermore, we leverage transformer architecture to generate co-speech facial and gesture actions, enabling real-time conversation with the generated character. These technologies have been integrated into our conversational AI avatar products.
comment: Technical Report
♻ ☆ Multi-modal and Multi-scale Spatial Environment Understanding for Immersive Visual Text-to-Speech AAAI'2025
Visual Text-to-Speech (VTTS) aims to take the environmental image as the prompt to synthesize the reverberant speech for the spoken content. The challenge of this task lies in understanding the spatial environment from the image. Many attempts have been made to extract global spatial visual information from the RGB space of an spatial image. However, local and depth image information are crucial for understanding the spatial environment, which previous works have ignored. To address the issues, we propose a novel multi-modal and multi-scale spatial environment understanding scheme to achieve immersive VTTS, termed M2SE-VTTS. The multi-modal aims to take both the RGB and Depth spaces of the spatial image to learn more comprehensive spatial information, and the multi-scale seeks to model the local and global spatial knowledge simultaneously. Specifically, we first split the RGB and Depth images into patches and adopt the Gemini-generated environment captions to guide the local spatial understanding. After that, the multi-modal and multi-scale features are integrated by the local-aware global spatial understanding. In this way, M2SE-VTTS effectively models the interactions between local and global spatial contexts in the multi-modal spatial environment. Objective and subjective evaluations suggest that our model outperforms the advanced baselines in environmental speech generation. The code and audio samples are available at: https://github.com/AI-S2-Lab/M2SE-VTTS.
comment: 9 pages,2 figures, Accepted by AAAI'2025
♻ ☆ Multi-Context Temporal Consistent Modeling for Referring Video Object Segmentation ICASSP 2025
Referring video object segmentation aims to segment objects within a video corresponding to a given text description. Existing transformer-based temporal modeling approaches face challenges related to query inconsistency and the limited consideration of context. Query inconsistency produces unstable masks of different objects in the middle of the video. The limited consideration of context leads to the segmentation of incorrect objects by failing to adequately account for the relationship between the given text and instances. To address these issues, we propose the Multi-context Temporal Consistency Module (MTCM), which consists of an Aligner and a Multi-Context Enhancer (MCE). The Aligner removes noise from queries and aligns them to achieve query consistency. The MCE predicts text-relevant queries by considering multi-context. We applied MTCM to four different models, increasing performance across all of them, particularly achieving 47.6 J&F on the MeViS. Code is available at https://github.com/Choi58/MTCM.
comment: Comment: Accepted to ICASSP 2025
♻ ☆ Adaptive Noise-Tolerant Network for Image Segmentation
Unlike image classification and annotation, for which deep network models have achieved dominating superior performances compared to traditional computer vision algorithms, deep learning for automatic image segmentation still faces critical challenges. One of such hurdles is to obtain ground-truth segmentations as the training labels for deep network training. Especially when we study biomedical images, such as histopathological images (histo-images), it is unrealistic to ask for manual segmentation labels as the ground truth for training due to the fine image resolution as well as the large image size and complexity. In this paper, instead of relying on clean segmentation labels, we study whether and how integrating imperfect or noisy segmentation results from off-the-shelf segmentation algorithms may help achieve better segmentation results through a new Adaptive Noise-Tolerant Network (ANTN) model. We extend the noisy label deep learning to image segmentation with two novel aspects: (1) multiple noisy labels can be integrated into one deep learning model; (2) noisy segmentation modeling, including probabilistic parameters, is adaptive, depending on the given testing image appearance. Implementation of the new ANTN model on both the synthetic data and real-world histo-images demonstrates its effectiveness and superiority over off-the-shelf and other existing deep-learning-based image segmentation algorithms.
♻ ☆ Efficient Long Video Tokenization via Coordinate-based Patch Reconstruction
Efficient tokenization of videos remains a challenge in training vision models that can process long videos. One promising direction is to develop a tokenizer that can encode long video clips, as it would enable the tokenizer to leverage the temporal coherence of videos better for tokenization. However, training existing tokenizers on long videos often incurs a huge training cost as they are trained to reconstruct all the frames at once. In this paper, we introduce CoordTok, a video tokenizer that learns a mapping from coordinate-based representations to the corresponding patches of input videos, inspired by recent advances in 3D generative models. In particular, CoordTok encodes a video into factorized triplane representations and reconstructs patches that correspond to randomly sampled $(x,y,t)$ coordinates. This allows for training large tokenizer models directly on long videos without requiring excessive training resources. Our experiments show that CoordTok can drastically reduce the number of tokens for encoding long video clips. For instance, CoordTok can encode a 128-frame video with 128$\times$128 resolution into 1280 tokens, while baselines need 6144 or 8192 tokens to achieve similar reconstruction quality. We further show that this efficient video tokenization enables memory-efficient training of a diffusion transformer that can generate 128 frames at once.
comment: Code is available on the project webpage: https://huiwon-jang.github.io/coordtok/
♻ ☆ A Unifying Information-theoretic Perspective on Evaluating Generative Models
Considering the difficulty of interpreting generative model output, there is significant current research focused on determining meaningful evaluation metrics. Several recent approaches utilize "precision" and "recall," borrowed from the classification domain, to individually quantify the output fidelity (realism) and output diversity (representation of the real data variation), respectively. With the increase in metric proposals, there is a need for a unifying perspective, allowing for easier comparison and clearer explanation of their benefits and drawbacks. To this end, we unify a class of kth-nearest-neighbors (kNN)-based metrics under an information-theoretic lens using approaches from kNN density estimation. Additionally, we propose a tri-dimensional metric composed of Precision Cross-Entropy (PCE), Recall Cross-Entropy (RCE), and Recall Entropy (RE), which separately measure fidelity and two distinct aspects of diversity, inter- and intra-class. Our domain-agnostic metric, derived from the information-theoretic concepts of entropy and cross-entropy, can be dissected for both sample- and mode-level analysis. Our detailed experimental results demonstrate the sensitivity of our metric components to their respective qualities and reveal undesirable behaviors of other metrics.
♻ ☆ Enhancing Skin Disease Diagnosis: Interpretable Visual Concept Discovery with SAM WACV 2025
Current AI-assisted skin image diagnosis has achieved dermatologist-level performance in classifying skin cancer, driven by rapid advancements in deep learning architectures. However, unlike traditional vision tasks, skin images in general present unique challenges due to the limited availability of well-annotated datasets, complex variations in conditions, and the necessity for detailed interpretations to ensure patient safety. Previous segmentation methods have sought to reduce image noise and enhance diagnostic performance, but these techniques require fine-grained, pixel-level ground truth masks for training. In contrast, with the rise of foundation models, the Segment Anything Model (SAM) has been introduced to facilitate promptable segmentation, enabling the automation of the segmentation process with simple yet effective prompts. Efforts applying SAM predominantly focus on dermatoscopy images, which present more easily identifiable lesion boundaries than clinical photos taken with smartphones. This limitation constrains the practicality of these approaches to real-world applications. To overcome the challenges posed by noisy clinical photos acquired via non-standardized protocols and to improve diagnostic accessibility, we propose a novel Cross-Attentive Fusion framework for interpretable skin lesion diagnosis. Our method leverages SAM to generate visual concepts for skin diseases using prompts, integrating local visual concepts with global image features to enhance model performance. Extensive evaluation on two skin disease datasets demonstrates our proposed method's effectiveness on lesion diagnosis and interpretability.
comment: This paper is accepted by WACV 2025
♻ ☆ GauFRe: Gaussian Deformation Fields for Real-time Dynamic Novel View Synthesis WACV 2025
We propose a method that achieves state-of-the-art rendering quality and efficiency on monocular dynamic scene reconstruction using deformable 3D Gaussians. Implicit deformable representations commonly model motion with a canonical space and time-dependent backward-warping deformation field. Our method, GauFRe, uses a forward-warping deformation to explicitly model non-rigid transformations of scene geometry. Specifically, we propose a template set of 3D Gaussians residing in a canonical space, and a time-dependent forward-warping deformation field to model dynamic objects. Additionally, we tailor a 3D Gaussian-specific static component supported by an inductive bias-aware initialization approach which allows the deformation field to focus on moving scene regions, improving the rendering of complex real-world motion. The differentiable pipeline is optimized end-to-end with a self-supervised rendering loss. Experiments show our method achieves competitive results and higher efficiency than both previous state-of-the-art NeRF and Gaussian-based methods. For real-world scenes, GauFRe can train in ~20 mins and offer 96 FPS real-time rendering on an RTX 3090 GPU. Project website: https://lynl7130.github.io/gaufre/index.html
comment: WACV 2025. 11 pages, 8 figures, 5 tables
♻ ☆ Key-Exchange Convolutional Auto-Encoder for Data Augmentation in Early Knee Osteoarthritis Detection
Knee Osteoarthritis (KOA) is a common musculoskeletal condition that significantly affects mobility and quality of life, particularly in elderly populations. However, training deep learning models for early KOA classification is often hampered by the limited availability of annotated medical datasets, owing to the high costs and labour-intensive nature of data labelling. Traditional data augmentation techniques, while useful, rely on simple transformations and fail to introduce sufficient diversity into the dataset. To address these challenges, we propose the Key-Exchange Convolutional Auto-Encoder (KECAE) as an innovative Artificial Intelligence (AI)-based data augmentation strategy for early KOA classification. Our model employs a convolutional autoencoder with a novel key-exchange mechanism that generates synthetic images by selectively exchanging key pathological features between X-ray images, which not only diversifies the dataset but also ensures the clinical validity of the augmented data. A hybrid loss function is introduced to supervise feature learning and reconstruction, integrating multiple components, including reconstruction, supervision, and feature separation losses. Experimental results demonstrate that the KECAE-generated data significantly improve the performance of KOA classification models, with accuracy gains of up to 1.98% across various standard and state-of-the-art architectures. Furthermore, a clinical validation study involving expert radiologists confirms the anatomical plausibility and diagnostic realism of the synthetic outputs. These findings highlight the potential of KECAE as a robust tool for augmenting medical datasets in early KOA detection.
♻ ☆ Human Activity Recognition in an Open World
Managing novelty in perception-based human activity recognition (HAR) is critical in realistic settings to improve task performance over time and ensure solution generalization outside of prior seen samples. Novelty manifests in HAR as unseen samples, activities, objects, environments, and sensor changes, among other ways. Novelty may be task-relevant, such as a new class or new features, or task-irrelevant resulting in nuisance novelty, such as never before seen noise, blur, or distorted video recordings. To perform HAR optimally, algorithmic solutions must be tolerant to nuisance novelty, and learn over time in the face of novelty. This paper 1) formalizes the definition of novelty in HAR building upon the prior definition of novelty in classification tasks, 2) proposes an incremental open world learning (OWL) protocol and applies it to the Kinetics datasets to generate a new benchmark KOWL-718, 3) analyzes the performance of current state-of-the-art HAR models when novelty is introduced over time, 4) provides a containerized and packaged pipeline for reproducing the OWL protocol and for modifying for any future updates to Kinetics. The experimental analysis includes an ablation study of how the different models perform under various conditions as annotated by Kinetics-AVA. The protocol as an algorithm for reproducing experiments using the KOWL-718 benchmark will be publicly released with code and containers at https://github.com/prijatelj/human-activity-recognition-in-an-open-world. The code may be used to analyze different annotations and subsets of the Kinetics datasets in an incremental open world fashion, as well as be extended as further updates to Kinetics are released.
comment: 37 pages, 16 figures, 3 tables. Published in JAIR 81 on Dec 20, 2024. All author affiliations are from during the paper's original funded work. Updated info and current emails are provided in this version's first page
♻ ☆ Confidence-Driven Deep Learning Framework for Early Detection of Knee Osteoarthritis
Knee Osteoarthritis (KOA) is a prevalent musculoskeletal disorder that severely impacts mobility and quality of life, particularly among older adults. Its diagnosis often relies on subjective assessments using the Kellgren-Lawrence (KL) grading system, leading to variability in clinical evaluations. To address these challenges, we propose a confidence-driven deep learning framework for early KOA detection, focusing on distinguishing KL-0 and KL-2 stages. The Siamese-based framework integrates a novel multi-level feature extraction architecture with a hybrid loss strategy. Specifically, multi-level Global Average Pooling (GAP) layers are employed to extract features from varying network depths, ensuring comprehensive feature representation, while the hybrid loss strategy partitions training samples into high-, medium-, and low-confidence subsets. Tailored loss functions are applied to improve model robustness and effectively handle uncertainty in annotations. Experimental results on the Osteoarthritis Initiative (OAI) dataset demonstrate that the proposed framework achieves competitive accuracy, sensitivity, and specificity, comparable to those of expert radiologists. Cohen's kappa values (k > 0.85)) confirm substantial agreement, while McNemar's test (p > 0.05) indicates no statistically significant differences between the model and radiologists. Additionally, Confidence distribution analysis reveals that the model emulates radiologists' decision-making patterns. These findings highlight the potential of the proposed approach to serve as an auxiliary diagnostic tool, enhancing early KOA detection and reducing clinical workload.
♻ ☆ Enhancing Novel Object Detection via Cooperative Foundational Models WACV 2025
In this work, we address the challenging and emergent problem of novel object detection (NOD), focusing on the accurate detection of both known and novel object categories during inference. Traditional object detection algorithms are inherently closed-set, limiting their capability to handle NOD. We present a novel approach to transform existing closed-set detectors into open-set detectors. This transformation is achieved by leveraging the complementary strengths of pre-trained foundational models, specifically CLIP and SAM, through our cooperative mechanism. Furthermore, by integrating this mechanism with state-of-the-art open-set detectors such as GDINO, we establish new benchmarks in object detection performance. Our method achieves 17.42 mAP in novel object detection and 42.08 mAP for known objects on the challenging LVIS dataset. Adapting our approach to the COCO OVD split, we surpass the current state-of-the-art by a margin of 7.2 $ \text{AP}_{50} $ for novel classes. Our code is available at https://rohit901.github.io/coop-foundation-models/ .
comment: Accepted at WACV 2025
♻ ☆ MVTamperBench: Evaluating Robustness of Vision-Language Models
Recent advancements in Vision-Language Models (VLMs) have enabled significant progress in complex video understanding tasks. However, their robustness to real-world manipulations remains underexplored, limiting their reliability in critical applications. To address this gap, we introduce MVTamperBench, a comprehensive benchmark designed to evaluate VLM's resilience to video tampering effects, including rotation, dropping, masking, substitution, and repetition. By systematically assessing state-of-the-art models, MVTamperBench reveals substantial variability in robustness, with models like InternVL2-8B achieving high performance, while others, such as Llama-VILA1.5-8B, exhibit severe vulnerabilities. To foster broader adoption and reproducibility, MVTamperBench is integrated into VLMEvalKit, a modular evaluation toolkit, enabling streamlined testing and facilitating advancements in model robustness. Our benchmark represents a critical step towards developing tamper-resilient VLMs, ensuring their dependability in real-world scenarios. Project Page: https://amitbcp.github.io/MVTamperBench/
Computational Geometry 1
☆ Is magnitude 'generically continuous' for finite metric spaces?
Magnitude is a real-valued invariant of metric spaces which, in the finite setting, can be understood as recording the 'effective number of points' in a space as the scale of the metric varies. Motivated by applications in topological data analysis, this paper investigates the stability of magnitude: its continuity properties with respect to the Gromov-Hausdorff topology. We show that magnitude is nowhere continuous on the Gromov-Hausdorff space of finite metric spaces. Yet, we find evidence to suggest that it may be 'generically continuous', in the sense that generic Gromov-Hausdorff limits are preserved by magnitude. We make the case that, in fact, 'generic stability' is what matters for applicability.
comment: 18 pages, 3 figures
Computation and Language 70
☆ Multimodal LLMs Can Reason about Aesthetics in Zero-Shot
We present the first study on how Multimodal LLMs' (MLLMs) reasoning ability shall be elicited to evaluate the aesthetics of artworks. To facilitate this investigation, we construct MM-StyleBench, a novel high-quality dataset for benchmarking artistic stylization. We then develop a principled method for human preference modeling and perform a systematic correlation analysis between MLLMs' responses and human preference. Our experiments reveal an inherent hallucination issue of MLLMs in art evaluation, associated with response subjectivity. ArtCoT is proposed, demonstrating that art-specific task decomposition and the use of concrete language boost MLLMs' reasoning ability for aesthetics. Our findings offer valuable insights into MLLMs for art and can benefit a wide range of downstream applications, such as style transfer and artistic image generation. Code available at https://github.com/songrise/MLLM4Art.
comment: WIP, Homepage https://github.com/songrise/MLLM4Art
☆ Aegis2.0: A Diverse AI Safety Dataset and Risks Taxonomy for Alignment of LLM Guardrails
As Large Language Models (LLMs) and generative AI become increasingly widespread, concerns about content safety have grown in parallel. Currently, there is a clear lack of high-quality, human-annotated datasets that address the full spectrum of LLM-related safety risks and are usable for commercial applications. To bridge this gap, we propose a comprehensive and adaptable taxonomy for categorizing safety risks, structured into 12 top-level hazard categories with an extension to 9 fine-grained subcategories. This taxonomy is designed to meet the diverse requirements of downstream users, offering more granular and flexible tools for managing various risk types. Using a hybrid data generation pipeline that combines human annotations with a multi-LLM "jury" system to assess the safety of responses, we obtain Aegis 2.0, a carefully curated collection of 34,248 samples of human-LLM interactions, annotated according to our proposed taxonomy. To validate its effectiveness, we demonstrate that several lightweight models, trained using parameter-efficient techniques on Aegis 2.0, achieve performance competitive with leading safety models fully fine-tuned on much larger, non-commercial datasets. In addition, we introduce a novel training blend that combines safety with topic following data.This approach enhances the adaptability of guard models, enabling them to generalize to new risk categories defined during inference. We plan to open-source Aegis 2.0 data and models to the research community to aid in the safety guardrailing of LLMs.
comment: arXiv admin note: text overlap with arXiv:2404.05993
☆ Personality Modeling for Persuasion of Misinformation using AI Agent
The proliferation of misinformation on social media platforms has highlighted the need to understand how individual personality traits influence susceptibility to and propagation of misinformation. This study employs an innovative agent-based modeling approach to investigate the relationship between personality traits and misinformation dynamics. Using six AI agents embodying different dimensions of the Big Five personality traits (Extraversion, Agreeableness, and Neuroticism), we simulated interactions across six diverse misinformation topics. The experiment, implemented through the AgentScope framework using the GLM-4-Flash model, generated 90 unique interactions, revealing complex patterns in how personality combinations affect persuasion and resistance to misinformation. Our findings demonstrate that analytical and critical personality traits enhance effectiveness in evidence-based discussions, while non-aggressive persuasion strategies show unexpected success in misinformation correction. Notably, agents with critical traits achieved a 59.4% success rate in HIV-related misinformation discussions, while those employing non-aggressive approaches maintained consistent persuasion rates above 40% across different personality combinations. The study also revealed a non-transitive pattern in persuasion effectiveness, challenging conventional assumptions about personality-based influence. These results provide crucial insights for developing personality-aware interventions in digital environments and suggest that effective misinformation countermeasures should prioritize emotional connection and trust-building over confrontational approaches. The findings contribute to both theoretical understanding of personality-misinformation dynamics and practical strategies for combating misinformation in social media contexts.
☆ Learning to Extract Cross-Domain Aspects and Understanding Sentiments Using Large Language Models
Aspect-based sentiment analysis (ASBA) is a refined approach to sentiment analysis that aims to extract and classify sentiments based on specific aspects or features of a product, service, or entity. Unlike traditional sentiment analysis, which assigns a general sentiment score to entire reviews or texts, ABSA focuses on breaking down the text into individual components or aspects (e.g., quality, price, service) and evaluating the sentiment towards each. This allows for a more granular level of understanding of customer opinions, enabling businesses to pinpoint specific areas of strength and improvement. The process involves several key steps, including aspect extraction, sentiment classification, and aspect-level sentiment aggregation for a review paragraph or any other form that the users have provided. ABSA has significant applications in areas such as product reviews, social media monitoring, customer feedback analysis, and market research. By leveraging techniques from natural language processing (NLP) and machine learning, ABSA facilitates the extraction of valuable insights, enabling companies to make data-driven decisions that enhance customer satisfaction and optimize offerings. As ABSA evolves, it holds the potential to greatly improve personalized customer experiences by providing a deeper understanding of sentiment across various product aspects. In this work, we have analyzed the strength of LLMs for a complete cross-domain aspect-based sentiment analysis with the aim of defining the framework for certain products and using it for other similar situations. We argue that it is possible to that at an effectiveness of 92\% accuracy for the Aspect Based Sentiment Analysis dataset of SemEval-2015 Task 12.
☆ Applying General Turn-taking Models to Conversational Human-Robot Interaction
Turn-taking is a fundamental aspect of conversation, but current Human-Robot Interaction (HRI) systems often rely on simplistic, silence-based models, leading to unnatural pauses and interruptions. This paper investigates, for the first time, the application of general turn-taking models, specifically TurnGPT and Voice Activity Projection (VAP), to improve conversational dynamics in HRI. These models are trained on human-human dialogue data using self-supervised learning objectives, without requiring domain-specific fine-tuning. We propose methods for using these models in tandem to predict when a robot should begin preparing responses, take turns, and handle potential interruptions. We evaluated the proposed system in a within-subject study against a traditional baseline system, using the Furhat robot with 39 adults in a conversational setting, in combination with a large language model for autonomous response generation. The results show that participants significantly prefer the proposed system, and it significantly reduces response delays and interruptions.
comment: Accepted at HRI 2025 (the IEEE/ACM International Conference on Human-Robot Interaction)
☆ Disentangling Exploration of Large Language Models by Optimal Exploitation
Exploration is a crucial skill for self-improvement and open-ended problem-solving. However, it remains uncertain whether large language models can effectively explore the state-space. Existing evaluations predominantly focus on the trade-off between exploration and exploitation, often assessed in multi-armed bandit problems. In contrast, this work isolates exploration as the sole objective, tasking the agent with delivering information that enhances future returns. For the evaluation, we propose to decompose missing rewards into exploration and exploitation components by measuring the optimal achievable return for the states already explored. Our experiments with various LLMs reveal that most models struggle to sufficiently explore the state-space and that weak exploration is insufficient. We observe a positive correlation between model size and exploration performance, with larger models demonstrating superior capabilities. Furthermore, we show that our decomposition provides insights into differences in behaviors driven by agent instructions during prompt engineering, offering a valuable tool for refining LLM performance in exploratory tasks.
☆ GenAI Content Detection Task 3: Cross-Domain Machine-Generated Text Detection Challenge COLING 2025
Recently there have been many shared tasks targeting the detection of generated text from Large Language Models (LLMs). However, these shared tasks tend to focus either on cases where text is limited to one particular domain or cases where text can be from many domains, some of which may not be seen during test time. In this shared task, using the newly released RAID benchmark, we aim to answer whether or not models can detect generated text from a large, yet fixed, number of domains and LLMs, all of which are seen during training. Over the course of three months, our task was attempted by 9 teams with 23 detector submissions. We find that multiple participants were able to obtain accuracies of over 99% on machine-generated text from RAID while maintaining a 5% False Positive Rate -- suggesting that detectors are able to robustly detect text from many domains and models simultaneously. We discuss potential interpretations of this result and provide directions for future research.
comment: COLING 2025
☆ ToMATO: Verbalizing the Mental States of Role-Playing LLMs for Benchmarking Theory of Mind AAAI 2025
Existing Theory of Mind (ToM) benchmarks diverge from real-world scenarios in three aspects: 1) they assess a limited range of mental states such as beliefs, 2) false beliefs are not comprehensively explored, and 3) the diverse personality traits of characters are overlooked. To address these challenges, we introduce ToMATO, a new ToM benchmark formulated as multiple-choice QA over conversations. ToMATO is generated via LLM-LLM conversations featuring information asymmetry. By employing a prompting method that requires role-playing LLMs to verbalize their thoughts before each utterance, we capture both first- and second-order mental states across five categories: belief, intention, desire, emotion, and knowledge. These verbalized thoughts serve as answers to questions designed to assess the mental states of characters within conversations. Furthermore, the information asymmetry introduced by hiding thoughts from others induces the generation of false beliefs about various mental states. Assigning distinct personality traits to LLMs further diversifies both utterances and thoughts. ToMATO consists of 5.4k questions, 753 conversations, and 15 personality trait patterns. Our analysis shows that this dataset construction approach frequently generates false beliefs due to the information asymmetry between role-playing LLMs, and effectively reflects diverse personalities. We evaluate nine LLMs on ToMATO and find that even GPT-4o mini lags behind human performance, especially in understanding false beliefs, and lacks robustness to various personality traits.
comment: Accepted by AAAI 2025
☆ MMDocIR: Benchmarking Multi-Modal Retrieval for Long Documents
Multi-modal document retrieval is designed to identify and retrieve various forms of multi-modal content, such as figures, tables, charts, and layout information from extensive documents. Despite its significance, there is a notable lack of a robust benchmark to effectively evaluate the performance of systems in multi-modal document retrieval. To address this gap, this work introduces a new benchmark, named as MMDocIR, encompassing two distinct tasks: page-level and layout-level retrieval. The former focuses on localizing the most relevant pages within a long document, while the latter targets the detection of specific layouts, offering a more fine-grained granularity than whole-page analysis. A layout can refer to a variety of elements such as textual paragraphs, equations, figures, tables, or charts. The MMDocIR benchmark comprises a rich dataset featuring expertly annotated labels for 1,685 questions and bootstrapped labels for 173,843 questions, making it a pivotal resource for advancing multi-modal document retrieval for both training and evaluation. Through rigorous experiments, we reveal that (i) visual retrievers significantly outperform their text counterparts, (ii) MMDocIR train set can effectively benefit the training process of multi-modal document retrieval and (iii) text retrievers leveraging on VLM-text perform much better than those using OCR-text. These findings underscores the potential advantages of integrating visual elements for multi-modal document retrieval.
comment: https://huggingface.co/MMDocIR
☆ SAIF: A Comprehensive Framework for Evaluating the Risks of Generative AI in the Public Sector AAAI
The rapid adoption of generative AI in the public sector, encompassing diverse applications ranging from automated public assistance to welfare services and immigration processes, highlights its transformative potential while underscoring the pressing need for thorough risk assessments. Despite its growing presence, evaluations of risks associated with AI-driven systems in the public sector remain insufficiently explored. Building upon an established taxonomy of AI risks derived from diverse government policies and corporate guidelines, we investigate the critical risks posed by generative AI in the public sector while extending the scope to account for its multimodal capabilities. In addition, we propose a Systematic dAta generatIon Framework for evaluating the risks of generative AI (SAIF). SAIF involves four key stages: breaking down risks, designing scenarios, applying jailbreak methods, and exploring prompt types. It ensures the systematic and consistent generation of prompt data, facilitating a comprehensive evaluation while providing a solid foundation for mitigating the risks. Furthermore, SAIF is designed to accommodate emerging jailbreak methods and evolving prompt types, thereby enabling effective responses to unforeseen risk scenarios. We believe that this study can play a crucial role in fostering the safe and responsible integration of generative AI into the public sector.
comment: 6 pages, 2 figures, 1 tables. AI for Public Missions (AIPM) Workshop at the 39th AAAI Conference on Artificial Intelligence (AAAI 2025)
☆ Enhanced Large Language Models for Effective Screening of Depression and Anxiety
Depressive and anxiety disorders are widespread, necessitating timely identification and management. Recent advances in Large Language Models (LLMs) offer potential solutions, yet high costs and ethical concerns about training data remain challenges. This paper introduces a pipeline for synthesizing clinical interviews, resulting in 1,157 interactive dialogues (PsyInterview), and presents EmoScan, an LLM-based emotional disorder screening system. EmoScan distinguishes between coarse (e.g., anxiety or depressive disorders) and fine disorders (e.g., major depressive disorders) and conducts high-quality interviews. Evaluations showed that EmoScan exceeded the performance of base models and other LLMs like GPT-4 in screening emotional disorders (F1-score=0.7467). It also delivers superior explanations (BERTScore=0.9408) and demonstrates robust generalizability (F1-score of 0.67 on an external dataset). Furthermore, EmoScan outperforms baselines in interviewing skills, as validated by automated ratings and human evaluations. This work highlights the importance of scalable data-generative pipelines for developing effective mental health LLM tools.
☆ Expanding Vietnamese SentiWordNet to Improve Performance of Vietnamese Sentiment Analysis Models
Sentiment analysis is one of the most crucial tasks in Natural Language Processing (NLP), involving the training of machine learning models to classify text based on the polarity of opinions. Pre-trained Language Models (PLMs) can be applied to downstream tasks through fine-tuning, eliminating the need to train the model from scratch. Specifically, PLMs have been employed for Sentiment Analysis, a process that involves detecting, analyzing, and extracting the polarity of text sentiments. Numerous models have been proposed to address this task, with pre-trained PhoBERT-V2 models standing out as the state-of-the-art language models for Vietnamese. The PhoBERT-V2 pre-training approach is based on RoBERTa, optimizing the BERT pre-training method for more robust performance. In this paper, we introduce a novel approach that combines PhoBERT-V2 and SentiWordnet for Sentiment Analysis of Vietnamese reviews. Our proposed model utilizes PhoBERT-V2 for Vietnamese, offering a robust optimization for the prominent BERT model in the context of Vietnamese language, and leverages SentiWordNet, a lexical resource explicitly designed to support sentiment classification applications. Experimental results on the VLSP 2016 and AIVIVN 2019 datasets demonstrate that our sentiment analysis system has achieved excellent performance in comparison to other models.
☆ The Inherent Limits of Pretrained LLMs: The Unexpected Convergence of Instruction Tuning and In-Context Learning Capabilities
Large Language Models (LLMs), trained on extensive web-scale corpora, have demonstrated remarkable abilities across diverse tasks, especially as they are scaled up. Nevertheless, even state-of-the-art models struggle in certain cases, sometimes failing at problems solvable by young children, indicating that traditional notions of task complexity are insufficient for explaining LLM capabilities. However, exploring LLM capabilities is complicated by the fact that most widely-used models are also "instruction-tuned" to respond appropriately to prompts. With the goal of disentangling the factors influencing LLM performance, we investigate whether instruction-tuned models possess fundamentally different capabilities from base models that are prompted using in-context examples. Through extensive experiments across various model families, scales and task types, which included instruction tuning 90 different LLMs, we demonstrate that the performance of instruction-tuned models is significantly correlated with the in-context performance of their base counterparts. By clarifying what instruction-tuning contributes, we extend prior research into in-context learning, which suggests that base models use priors from pretraining data to solve tasks. Specifically, we extend this understanding to instruction-tuned models, suggesting that their pretraining data similarly sets a limiting boundary on the tasks they can solve, with the added influence of the instruction-tuning dataset.
comment: The code for this paper is available at: https://github.com/UKPLab/arxiv2025-inherent-limits-plms
☆ Deep Learning-Based Feature Fusion for Emotion Analysis and Suicide Risk Differentiation in Chinese Psychological Support Hotlines
Mental health is a critical global public health issue, and psychological support hotlines play a pivotal role in providing mental health assistance and identifying suicide risks at an early stage. However, the emotional expressions conveyed during these calls remain underexplored in current research. This study introduces a method that combines pitch acoustic features with deep learning-based features to analyze and understand emotions expressed during hotline interactions. Using data from China's largest psychological support hotline, our method achieved an F1-score of 79.13% for negative binary emotion classification.Additionally, the proposed approach was validated on an open dataset for multi-class emotion classification,where it demonstrated better performance compared to the state-of-the-art methods. To explore its clinical relevance, we applied the model to analysis the frequency of negative emotions and the rate of emotional change in the conversation, comparing 46 subjects with suicidal behavior to those without. While the suicidal group exhibited more frequent emotional changes than the non-suicidal group, the difference was not statistically significant.Importantly, our findings suggest that emotional fluctuation intensity and frequency could serve as novel features for psychological assessment scales and suicide risk prediction.The proposed method provides valuable insights into emotional dynamics and has the potential to advance early intervention and improve suicide prevention strategies through integration with clinical tools and assessments The source code is publicly available at https://github.com/Sco-field/Speechemotionrecognition/tree/main.
☆ Knowledge Graph-based Retrieval-Augmented Generation for Schema Matching
Traditional similarity-based schema matching methods are incapable of resolving semantic ambiguities and conflicts in domain-specific complex mapping scenarios due to missing commonsense and domain-specific knowledge. The hallucination problem of large language models (LLMs) also makes it challenging for LLM-based schema matching to address the above issues. Therefore, we propose a Knowledge Graph-based Retrieval-Augmented Generation model for Schema Matching, referred to as the KG-RAG4SM. In particular, KG-RAG4SM introduces novel vector-based, graph traversal-based, and query-based graph retrievals, as well as a hybrid approach and ranking schemes that identify the most relevant subgraphs from external large knowledge graphs (KGs). We showcase that KG-based retrieval-augmented LLMs are capable of generating more accurate results for complex matching cases without any re-training. Our experimental results show that KG-RAG4SM outperforms the LLM-based state-of-the-art (SOTA) methods (e.g., Jellyfish-8B) by 35.89% and 30.50% in terms of precision and F1 score on the MIMIC dataset, respectively; KG-RAG4SM with GPT-4o-mini outperforms the pre-trained language model (PLM)-based SOTA methods (e.g., SMAT) by 69.20% and 21.97% in terms of precision and F1 score on the Synthea dataset, respectively. The results also demonstrate that our approach is more efficient in end-to-end schema matching, and scales to retrieve from large KGs. Our case studies on the dataset from the real-world schema matching scenario exhibit that the hallucination problem of LLMs for schema matching is well mitigated by our solution.
comment: Under Review
☆ MAGNET: Augmenting Generative Decoders with Representation Learning and Infilling Capabilities
While originally designed for unidirectional generative modeling, decoder-only large language models (LLMs) are increasingly being adapted for bidirectional modeling. However, unidirectional and bidirectional models are typically trained separately with distinct objectives (generation and representation learning, respectively). This separation overlooks the opportunity for developing a more versatile language model and for these objectives to complement each other. In this work, we introduce MAGNET, an adaptation of decoder-only LLMs that enhances their ability to generate robust representations and infill missing text spans, while preserving their knowledge and text generation capabilities. MAGNET employs three self-supervised training objectives and introduces an attention mechanism that combines bidirectional and causal attention, enabling unified training across all objectives. Our results demonstrate that LLMs adapted with MAGNET (1) surpass strong text encoders on token-level and sentence-level representation learning tasks, (2) generate contextually appropriate text infills by leveraging future context, (3) retain the ability for open-ended text generation without exhibiting repetition problem, and (4) preserve the knowledge gained by the LLM during pretraining.
☆ Reassessing the Role of Chain-of-Thought in Sentiment Analysis: Insights and Limitations
The relationship between language and thought remains an unresolved philosophical issue. Existing viewpoints can be broadly categorized into two schools: one asserting their independence, and another arguing that language constrains thought. In the context of large language models, this debate raises a crucial question: Does a language model's grasp of semantic meaning depend on thought processes? To explore this issue, we investigate whether reasoning techniques can facilitate semantic understanding. Specifically, we conceptualize thought as reasoning, employ chain-of-thought prompting as a reasoning technique, and examine its impact on sentiment analysis tasks. The experiments show that chain-of-thought has a minimal impact on sentiment analysis tasks. Both the standard and chain-of-thought prompts focus on aspect terms rather than sentiment in the generated content. Furthermore, counterfactual experiments reveal that the model's handling of sentiment tasks primarily depends on information from demonstrations. The experimental results support the first viewpoint.
☆ SWSC: Shared Weight for Similar Channel in LLM
Large language models (LLMs) have spurred development in multiple industries. However, the growing number of their parameters brings substantial storage and computing burdens, making it essential to explore model compression techniques for parameter reduction and easier deployment. We propose SWSC, an LLM compression method based on the concept of Shared Weight for Similar Channel. It uses the K-Means clustering algorithm to cluster model weights channel-by-channel, generating clusters with highly similar vectors within each. A representative vector from each cluster is selected to approximately replace all vectors in the cluster, significantly reducing the number of model weight parameters. However, approximate restoration will inevitably cause damage to the performance of the model. To tackle this issue, we perform singular value decomposition on the weight error values before and after compression and retain the larger singular values and their corresponding singular vectors to compensate for the accuracy. The experimental results show that our method can effectively ensure the performance of the compressed LLM even under low-precision conditions.
comment: 5pages, 3 figures, work in progress
☆ ViBidirectionMT-Eval: Machine Translation for Vietnamese-Chinese and Vietnamese-Lao language pair
This paper presents an results of the VLSP 2022-2023 Machine Translation Shared Tasks, focusing on Vietnamese-Chinese and Vietnamese-Lao machine translation. The tasks were organized as part of the 9th, 10th annual workshop on Vietnamese Language and Speech Processing (VLSP 2022, VLSP 2023). The objective of the shared task was to build machine translation systems, specifically targeting Vietnamese-Chinese and Vietnamese-Lao translation (corresponding to 4 translation directions). The submission were evaluated on 1,000 pairs for testing (news and general domains) using established metrics like BLEU [11] and SacreBLEU [12]. Additionally, system outputs also were evaluated with human judgment provided by experts in Chinese and Lao languages. These human assessments played a crucial role in ranking the performance of the machine translation models, ensuring a more comprehensive evaluation.
☆ Disjoint Processing Mechanisms of Hierarchical and Linear Grammars in Large Language Models
All natural languages are structured hierarchically. In humans, this structural restriction is neurologically coded: when two grammars are presented with identical vocabularies, brain areas responsible for language processing are only sensitive to hierarchical grammars. Using large language models (LLMs), we investigate whether such functionally distinct hierarchical processing regions can arise solely from exposure to large-scale language distributions. We generate inputs using English, Italian, Japanese, or nonce words, varying the underlying grammars to conform to either hierarchical or linear/positional rules. Using these grammars, we first observe that language models show distinct behaviors on hierarchical versus linearly structured inputs. Then, we find that the components responsible for processing hierarchical grammars are distinct from those that process linear grammars; we causally verify this in ablation experiments. Finally, we observe that hierarchy-selective components are also active on nonce grammars; this suggests that hierarchy sensitivity is not tied to meaning, nor in-distribution inputs.
☆ RLHS: Mitigating Misalignment in RLHF with Hindsight Simulation
Generative AI systems like foundation models (FMs) must align well with human values to ensure their behavior is helpful and trustworthy. While Reinforcement Learning from Human Feedback (RLHF) has shown promise for optimizing model performance using human judgments, existing RLHF pipelines predominantly rely on immediate feedback, which can fail to accurately reflect the downstream impact of an interaction on users' utility. We demonstrate that feedback based on evaluators' foresight estimates of downstream consequences systematically induces Goodhart's Law dynamics, incentivizing misaligned behaviors like sycophancy and deception and ultimately degrading user outcomes. To alleviate this, we propose decoupling evaluation from prediction by refocusing RLHF on hindsight feedback. Our theoretical analysis reveals that conditioning evaluator feedback on downstream observations mitigates misalignment and improves expected human utility, even when these observations are simulated by the AI system itself. To leverage this insight in a practical alignment algorithm, we introduce Reinforcement Learning from Hindsight Simulation (RLHS), which first simulates plausible consequences and then elicits feedback to assess what behaviors were genuinely beneficial in hindsight. We apply RLHS to two widely-employed online and offline preference optimization methods -- Proximal Policy Optimization (PPO) and Direct Preference Optimization (DPO) -- and show empirically that misalignment is significantly reduced with both methods. Through an online human user study, we show that RLHS consistently outperforms RLHF in helping users achieve their goals and earns higher satisfaction ratings, despite being trained solely with simulated hindsight feedback. These results underscore the importance of focusing on long-term consequences, even simulated ones, to mitigate misalignment in RLHF.
☆ Assessing the Alignment of FOL Closeness Metrics with Human Judgement
The recent successful paradigm of solving logical reasoning problems with tool-augmented large language models (LLMs) leverages translation of natural language statements into First-Order Logic~(FOL) and external theorem provers. However, the correctness of FOL statements, comprising operators and text predicates, often goes unverified due to the lack of a reliable evaluation metric for comparing generated and ground-truth FOLs. In this paper, we present a comprehensive study of sensitivity of existing metrics and their alignment with human judgement on FOL evaluation. Using ground-truth FOLs, we carefully designed various perturbations on the ground-truth to assess metric sensitivity. We sample FOL translation candidates for natural language statements and measure the ranking alignment between automatic metrics and human annotators. Our empirical findings highlight oversensitivity in the n-gram metric BLEU for text perturbations, the semantic graph metric Smatch++ for structural perturbations, and FOL metric for operator perturbation. We also observe a closer alignment between BertScore and human judgement. Additionally, we show that combining metrics enhances both alignment and sensitivity compared to using individual metrics.
comment: Code: https://github.com/RamyaKeerthy/AlignmentFOL
☆ Dynamic Knowledge Integration for Enhanced Vision-Language Reasoning
Large Vision-Language Models (LVLMs) have demonstrated impressive capabilities in multimodal tasks, but their performance is often constrained by the lack of external knowledge integration, limiting their ability to handle knowledge-intensive tasks such as visual question answering and reasoning. To address this challenge, we propose a novel method, Adaptive Knowledge-Guided Pretraining for Large Vision-Language Models (AKGP-LVLM), which dynamically incorporates structured and unstructured knowledge into LVLMs during pretraining and fine-tuning. Our approach employs a knowledge encoder to represent external knowledge, a retrieval mechanism to select task-relevant information, and a dynamic adaptor to align multimodal and knowledge representations effectively. We evaluate our method on four benchmark datasets, demonstrating significant performance improvements over state-of-the-art models. Furthermore, human evaluations highlight the superior correctness and relevance of our model's outputs. Extensive analyses confirm the robustness, efficiency, and scalability of AKGP-LVLM, making it a compelling solution for real-world knowledge-intensive tasks.
☆ LoRS: Efficient Low-Rank Adaptation for Sparse Large Language Model
Existing low-rank adaptation (LoRA) methods face challenges on sparse large language models (LLMs) due to the inability to maintain sparsity. Recent works introduced methods that maintain sparsity by augmenting LoRA techniques with additional masking mechanisms. Despite these successes, such approaches suffer from an increased memory and computation overhead, which affects efficiency of LoRA methods. In response to this limitation, we introduce LoRS, an innovative method designed to achieve both memory and computation efficiency when fine-tuning sparse LLMs. To mitigate the substantial memory and computation demands associated with preserving sparsity, our approach incorporates strategies of weight recompute and computational graph rearrangement. In addition, we also improve the effectiveness of LoRS through better adapter initialization. These innovations lead to a notable reduction in memory and computation consumption during the fine-tuning phase, all while achieving performance levels that outperform existing LoRA approaches.
comment: 12 pages, 4 figures
☆ What Limits LLM-based Human Simulation: LLMs or Our Design?
We argue that advancing LLM-based human simulation requires addressing both LLM's inherent limitations and simulation framework design challenges. Recent studies have revealed significant gaps between LLM-based human simulations and real-world observations, highlighting these dual challenges. To address these gaps, we present a comprehensive analysis of LLM limitations and our design issues, proposing targeted solutions for both aspects. Furthermore, we explore future directions that address both challenges simultaneously, particularly in data collection, LLM generation, and evaluation. To support further research in this field, we provide a curated collection of LLM-based human simulation resources.\footnote{https://github.com/Persdre/llm-human-simulation}
☆ Information Entropy Invariance: Enhancing Length Extrapolation in Attention Mechanisms
Improving the length extrapolation capabilities of Large Language Models (LLMs) remains a critical challenge in natural language processing. Many recent efforts have focused on modifying the scaled dot-product attention mechanism, and often introduce scaled temperatures without rigorous theoretical justification. To fill this gap, we introduce a novel approach based on information entropy invariance. We propose two new scaled temperatures to enhance length extrapolation. First, a training-free method InfoScale is designed for dot-product attention, and preserves focus on original tokens during length extrapolation by ensuring information entropy remains consistent. Second, we theoretically analyze the impact of scaling (CosScale) on cosine attention. Experimental data demonstrates that combining InfoScale and CosScale achieves state-of-the-art performance on the GAU-{\alpha} model with a context window extended to 64 times the training length, and outperforms seven existing methods. Our analysis reveals that significantly increasing CosScale approximates windowed attention, and highlights the significance of attention score dilution as a key challenge in long-range context handling. The code and data are available at https://github.com/HT-NEKO/InfoScale.
☆ Knowledge prompt chaining for semantic modeling
The task of building semantics for structured data such as CSV, JSON, and XML files is highly relevant in the knowledge representation field. Even though we have a vast of structured data on the internet, mapping them to domain ontologies to build semantics for them is still very challenging as it requires the construction model to understand and learn graph-structured knowledge. Otherwise, the task will require human beings' effort and cost. In this paper, we proposed a novel automatic semantic modeling framework: Knowledge Prompt Chaining. It can serialize the graph-structured knowledge and inject it into the LLMs properly in a Prompt Chaining architecture. Through this knowledge injection and prompting chaining, the model in our framework can learn the structure information and latent space of the graph and generate the semantic labels and semantic graphs following the chains' insturction naturally. Based on experimental results, our method achieves better performance than existing leading techniques, despite using reduced structured input data.
☆ Complexity Control Facilitates Reasoning-Based Compositional Generalization in Transformers
Transformers have demonstrated impressive capabilities across various tasks, yet their performance on compositional problems remains a subject of debate. In this study, we investigate the internal mechanisms underlying Transformers' behavior in compositional tasks. We find that complexity control strategies significantly influence whether the model learns primitive-level rules that generalize out-of-distribution (reasoning-based solutions) or relies solely on memorized mappings (memory-based solutions). By applying masking strategies to the model's information circuits and employing multiple complexity metrics, we reveal distinct internal working mechanisms associated with different solution types. Further analysis reveals that reasoning-based solutions exhibit a lower complexity bias, which aligns with the well-studied neuron condensation phenomenon. This lower complexity bias is hypothesized to be the key factor enabling these solutions to learn reasoning rules. We validate these conclusions across multiple real-world datasets, including image generation and natural language processing tasks, confirming the broad applicability of our findings.
comment: Mistakenly submitted as a replacement to 2405.05409v4
☆ Doc-Guided Sent2Sent++: A Sent2Sent++ Agent with Doc-Guided memory for Document-level Machine Translation
The field of artificial intelligence has witnessed significant advancements in natural language processing, largely attributed to the capabilities of Large Language Models (LLMs). These models form the backbone of Agents designed to address long-context dependencies, particularly in Document-level Machine Translation (DocMT). DocMT presents unique challenges, with quality, consistency, and fluency being the key metrics for evaluation. Existing approaches, such as Doc2Doc and Doc2Sent, either omit sentences or compromise fluency. This paper introduces Doc-Guided Sent2Sent++, an Agent that employs an incremental sentence-level forced decoding strategy \textbf{to ensure every sentence is translated while enhancing the fluency of adjacent sentences.} Our Agent leverages a Doc-Guided Memory, focusing solely on the summary and its translation, which we find to be an efficient approach to maintaining consistency. Through extensive testing across multiple languages and domains, we demonstrate that Sent2Sent++ outperforms other methods in terms of quality, consistency, and fluency. The results indicate that, our approach has achieved significant improvements in metrics such as s-COMET, d-COMET, LTCR-$1_f$, and document-level perplexity (d-ppl). The contributions of this paper include a detailed analysis of current DocMT research, the introduction of the Sent2Sent++ decoding method, the Doc-Guided Memory mechanism, and validation of its effectiveness across languages and domains.
☆ Adapting Whisper for Regional Dialects: Enhancing Public Services for Vulnerable Populations in the United Kingdom
We collect novel data in the public service domain to evaluate the capability of the state-of-the-art automatic speech recognition (ASR) models in capturing regional differences in accents in the United Kingdom (UK), specifically focusing on two accents from Scotland with distinct dialects. This study addresses real-world problems where biased ASR models can lead to miscommunication in public services, disadvantaging individuals with regional accents particularly those in vulnerable populations. We first examine the out-of-the-box performance of the Whisper large-v3 model on a baseline dataset and our data. We then explore the impact of fine-tuning Whisper on the performance in the two UK regions and investigate the effectiveness of existing model evaluation techniques for our real-world application through manual inspection of model errors. We observe that the Whisper model has a higher word error rate (WER) on our test datasets compared to the baseline data and fine-tuning on a given data improves performance on the test dataset with the same domain and accent. The fine-tuned models also appear to show improved performance when applied to the test data outside of the region it was trained on suggesting that fine-tuned models may be transferable within parts of the UK. Our manual analysis of model outputs reveals the benefits and drawbacks of using WER as an evaluation metric and fine-tuning to adapt to regional dialects.
☆ The Veln(ia)s is in the Details: Evaluating LLM Judgment on Latvian and Lithuanian Short Answer Matching
In this work, we address the challenge of evaluating large language models (LLMs) on the short answer matching task for Latvian and Lithuanian languages. We introduce novel datasets consisting of 502 Latvian and 690 Lithuanian question-answer pairs. For each question-answer pair, we generated matched and non-matched answers using a set of alteration rules specifically designed to introduce small but meaningful changes in the text. These generated answers serve as test cases to assess the ability of LLMs to detect subtle differences in matching of the original answers. A subset of the datasets was manually verified for quality and accuracy. Our results show that while larger LLMs, such as QWEN2.5 72b and LLaMa3.1 70b, demonstrate near-perfect performance in distinguishing matched and non-matched answers, smaller models show more variance. For instance, LLaMa3.1 8b and EuroLLM 9b benefited from few-shot examples, while Mistral Nemo 12b underperformed on detection of subtle text alteration, particularly in Lithuanian, even with additional examples. QWEN2.5 7b and Mistral 7b were able to obtain a strong and comparable performance to the larger 70b models in zero and few shot experiments. Moreover, the performance of Mistral 7b was weaker in few shot experiments.
☆ Evaluating GenAI for Simplifying Texts for Education: Improving Accuracy and Consistency for Enhanced Readability
Generative artificial intelligence (GenAI) holds great promise as a tool to support personalized learning. Teachers need tools to efficiently and effectively enhance content readability of educational texts so that they are matched to individual students reading levels, while retaining key details. Large Language Models (LLMs) show potential to fill this need, but previous research notes multiple shortcomings in current approaches. In this study, we introduced a generalized approach and metrics for the systematic evaluation of the accuracy and consistency in which LLMs, prompting techniques, and a novel multi-agent architecture to simplify sixty informational reading passages, reducing each from the twelfth grade level down to the eighth, sixth, and fourth grade levels. We calculated the degree to which each LLM and prompting technique accurately achieved the targeted grade level for each passage, percentage change in word count, and consistency in maintaining keywords and key phrases (semantic similarity). One-sample t-tests and multiple regression models revealed significant differences in the best performing LLM and prompt technique for each of the four metrics. Both LLMs and prompting techniques demonstrated variable utility in grade level accuracy and consistency of keywords and key phrases when attempting to level content down to the fourth grade reading level. These results demonstrate the promise of the application of LLMs for efficient and precise automated text simplification, the shortcomings of current models and prompting methods in attaining an ideal balance across various evaluation criteria, and a generalizable method to evaluate future systems.
comment: 64 pages, 9 tables, 6 figures, and supplemental materials
☆ VCRScore: Image captioning metric based on V\&L Transformers, CLIP, and precision-recall
Image captioning has become an essential Vision & Language research task. It is about predicting the most accurate caption given a specific image or video. The research community has achieved impressive results by continuously proposing new models and approaches to improve the overall model's performance. Nevertheless, despite increasing proposals, the performance metrics used to measure their advances have remained practically untouched through the years. A probe of that, nowadays metrics like BLEU, METEOR, CIDEr, and ROUGE are still very used, aside from more sophisticated metrics such as BertScore and ClipScore. Hence, it is essential to adjust how are measure the advances, limitations, and scopes of the new image captioning proposals, as well as to adapt new metrics to these new advanced image captioning approaches. This work proposes a new evaluation metric for the image captioning problem. To do that, first, it was generated a human-labeled dataset to assess to which degree the captions correlate with the image's content. Taking these human scores as ground truth, we propose a new metric, and compare it with several well-known metrics, from classical to newer ones. Outperformed results were also found, and interesting insights were presented and discussed.
comment: 28 pages
☆ Towards Multilingual LLM Evaluation for Baltic and Nordic languages: A study on Lithuanian History
In this work, we evaluated Lithuanian and general history knowledge of multilingual Large Language Models (LLMs) on a multiple-choice question-answering task. The models were tested on a dataset of Lithuanian national and general history questions translated into Baltic, Nordic, and other languages (English, Ukrainian, Arabic) to assess the knowledge sharing from culturally and historically connected groups. We evaluated GPT-4o, LLaMa3.1 8b and 70b, QWEN2.5 7b and 72b, Mistral Nemo 12b, LLaMa3 8b, Mistral 7b, LLaMa3.2 3b, and Nordic fine-tuned models (GPT-SW3 and LLaMa3 8b). Our results show that GPT-4o consistently outperformed all other models across language groups, with slightly better results for Baltic and Nordic languages. Larger open-source models like QWEN2.5 72b and LLaMa3.1 70b performed well but showed weaker alignment with Baltic languages. Smaller models (Mistral Nemo 12b, LLaMa3.2 3b, QWEN 7B, LLaMa3.1 8B, and LLaMa3 8b) demonstrated gaps with LT-related alignment with Baltic languages while performing better on Nordic and other languages. The Nordic fine-tuned models did not surpass multilingual models, indicating that shared cultural or historical context alone does not guarantee better performance.
☆ Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG
Large Language Models (LLMs) have revolutionized artificial intelligence (AI) by enabling human like text generation and natural language understanding. However, their reliance on static training data limits their ability to respond to dynamic, real time queries, resulting in outdated or inaccurate outputs. Retrieval Augmented Generation (RAG) has emerged as a solution, enhancing LLMs by integrating real time data retrieval to provide contextually relevant and up-to-date responses. Despite its promise, traditional RAG systems are constrained by static workflows and lack the adaptability required for multistep reasoning and complex task management. Agentic Retrieval-Augmented Generation (Agentic RAG) transcends these limitations by embedding autonomous AI agents into the RAG pipeline. These agents leverage agentic design patterns reflection, planning, tool use, and multiagent collaboration to dynamically manage retrieval strategies, iteratively refine contextual understanding, and adapt workflows to meet complex task requirements. This integration enables Agentic RAG systems to deliver unparalleled flexibility, scalability, and context awareness across diverse applications. This survey provides a comprehensive exploration of Agentic RAG, beginning with its foundational principles and the evolution of RAG paradigms. It presents a detailed taxonomy of Agentic RAG architectures, highlights key applications in industries such as healthcare, finance, and education, and examines practical implementation strategies. Additionally, it addresses challenges in scaling these systems, ensuring ethical decision making, and optimizing performance for real-world applications, while providing detailed insights into frameworks and tools for implementing Agentic RAG
☆ Benchmarking Robustness of Contrastive Learning Models for Medical Image-Report Retrieval AAAI 2025
Medical images and reports offer invaluable insights into patient health. The heterogeneity and complexity of these data hinder effective analysis. To bridge this gap, we investigate contrastive learning models for cross-domain retrieval, which associates medical images with their corresponding clinical reports. This study benchmarks the robustness of four state-of-the-art contrastive learning models: CLIP, CXR-RePaiR, MedCLIP, and CXR-CLIP. We introduce an occlusion retrieval task to evaluate model performance under varying levels of image corruption. Our findings reveal that all evaluated models are highly sensitive to out-of-distribution data, as evidenced by the proportional decrease in performance with increasing occlusion levels. While MedCLIP exhibits slightly more robustness, its overall performance remains significantly behind CXR-CLIP and CXR-RePaiR. CLIP, trained on a general-purpose dataset, struggles with medical image-report retrieval, highlighting the importance of domain-specific training data. The evaluation of this work suggests that more effort needs to be spent on improving the robustness of these models. By addressing these limitations, we can develop more reliable cross-domain retrieval models for medical applications.
comment: This work is accepted to AAAI 2025 Workshop -- the 9th International Workshop on Health Intelligence
☆ Multilingual LLMs Struggle to Link Orthography and Semantics in Bilingual Word Processing
Bilingual lexical processing is shaped by the complex interplay of phonological, orthographic, and semantic features of two languages within an integrated mental lexicon. In humans, this is evident in the ease with which cognate words - words similar in both orthographic form and meaning (e.g., blind, meaning "sightless" in both English and German) - are processed, compared to the challenges posed by interlingual homographs, which share orthographic form but differ in meaning (e.g., gift, meaning "present" in English but "poison" in German). We investigate how multilingual Large Language Models (LLMs) handle such phenomena, focusing on English-Spanish, English-French, and English-German cognates, non-cognate, and interlingual homographs. Specifically, we evaluate their ability to disambiguate meanings and make semantic judgments, both when these word types are presented in isolation or within sentence contexts. Our findings reveal that while certain LLMs demonstrate strong performance in recognizing cognates and non-cognates in isolation, they exhibit significant difficulty in disambiguating interlingual homographs, often performing below random baselines. This suggests LLMs tend to rely heavily on orthographic similarities rather than semantic understanding when interpreting interlingual homographs. Further, we find LLMs exhibit difficulty in retrieving word meanings, with performance in isolative disambiguation tasks having no correlation with semantic understanding. Finally, we study how the LLM processes interlingual homographs in incongruent sentences. We find models to opt for different strategies in understanding English and non-English homographs, highlighting a lack of a unified approach to handling cross-lingual ambiguities.
comment: Code available at: https://github.com/EshaanT/Bilingual_processing_LLMs
☆ Augmenting Human-Annotated Training Data with Large Language Model Generation and Distillation in Open-Response Assessment
Large Language Models (LLMs) like GPT-4o can help automate text classification tasks at low cost and scale. However, there are major concerns about the validity and reliability of LLM outputs. By contrast, human coding is generally more reliable but expensive to procure at scale. In this study, we propose a hybrid solution to leverage the strengths of both. We combine human-coded data and synthetic LLM-produced data to fine-tune a classical machine learning classifier, distilling both into a smaller BERT model. We evaluate our method on a human-coded test set as a validity measure for LLM output quality. In three experiments, we systematically vary LLM-generated samples' size, variety, and consistency, informed by best practices in LLM tuning. Our findings indicate that augmenting datasets with synthetic samples improves classifier performance, with optimal results achieved at an 80% synthetic to 20% human-coded data ratio. Lower temperature settings of 0.3, corresponding to less variability in LLM generations, produced more stable improvements but also limited model learning from augmented samples. In contrast, higher temperature settings (0.7 and above) introduced greater variability in performance estimates and, at times, lower performance. Hence, LLMs may produce more uniform output that classifiers overfit to earlier or produce more diverse output that runs the risk of deteriorating model performance through information irrelevant to the prediction task. Filtering out inconsistent synthetic samples did not enhance performance. We conclude that integrating human and LLM-generated data to improve text classification models in assessment offers a scalable solution that leverages both the accuracy of human coding and the variety of LLM outputs.
comment: Manuscript accepted to the Second Workshop on Generative AI for Learning Analytics (GenAI-LA) at LAK25
☆ SteLLA: A Structured Grading System Using LLMs with RAG
Large Language Models (LLMs) have shown strong general capabilities in many applications. However, how to make them reliable tools for some specific tasks such as automated short answer grading (ASAG) remains a challenge. We present SteLLA (Structured Grading System Using LLMs with RAG) in which a) Retrieval Augmented Generation (RAG) approach is used to empower LLMs specifically on the ASAG task by extracting structured information from the highly relevant and reliable external knowledge based on the instructor-provided reference answer and rubric, b) an LLM performs a structured and question-answering-based evaluation of student answers to provide analytical grades and feedback. A real-world dataset that contains students' answers in an exam was collected from a college-level Biology course. Experiments show that our proposed system can achieve substantial agreement with the human grader while providing break-down grades and feedback on all the knowledge points examined in the problem. A qualitative and error analysis of the feedback generated by GPT4 shows that GPT4 is good at capturing facts while may be prone to inferring too much implication from the given text in the grading task which provides insights into the usage of LLMs in the ASAG system.
☆ Decompose-ToM: Enhancing Theory of Mind Reasoning in Large Language Models through Simulation and Task Decomposition COLING 2025
Theory of Mind (ToM) is the ability to understand and reflect on the mental states of others. Although this capability is crucial for human interaction, testing on Large Language Models (LLMs) reveals that they possess only a rudimentary understanding of it. Although the most capable closed-source LLMs have come close to human performance on some ToM tasks, they still perform poorly on complex variations of the task that involve more structured reasoning. In this work, we utilize the concept of "pretend-play", or ``Simulation Theory'' from cognitive psychology to propose ``Decompose-ToM'': an LLM-based inference algorithm that improves model performance on complex ToM tasks. We recursively simulate user perspectives and decompose the ToM task into a simpler set of functions: subject identification, question-reframing, world model updation, and knowledge availability. We test the algorithm on higher-order ToM tasks and a task testing for ToM capabilities in a conversational setting, demonstrating that our approach shows significant improvement across models compared to baseline methods while requiring minimal prompt tuning across tasks and no additional model training.
comment: Accepted to COLING 2025
☆ Generative Visual Commonsense Answering and Explaining with Generative Scene Graph Constructing
Visual Commonsense Reasoning, which is regarded as one challenging task to pursue advanced visual scene comprehension, has been used to diagnose the reasoning ability of AI systems. However, reliable reasoning requires a good grasp of the scene's details. Existing work fails to effectively exploit the real-world object relationship information present within the scene, and instead overly relies on knowledge from training memory. Based on these observations, we propose a novel scene-graph-enhanced visual commonsense reasoning generation method named \textit{\textbf{G2}}, which first utilizes the image patches and LLMs to construct a location-free scene graph, and then answer and explain based on the scene graph's information. We also propose automatic scene graph filtering and selection strategies to absorb valuable scene graph information during training. Extensive experiments are conducted on the tasks and datasets of scene graph constructing and visual commonsense answering and explaining, respectively. Experimental results and ablation analysis demonstrate the effectiveness of our proposed framework.
♻ ☆ A General Framework for Inference-time Scaling and Steering of Diffusion Models
Diffusion models produce impressive results in modalities ranging from images and video to protein design and text. However, generating samples with user-specified properties remains a challenge. Recent research proposes fine-tuning models to maximize rewards that capture desired properties, but these methods require expensive training and are prone to mode collapse. In this work, we propose Feynman Kac (FK) steering, an inference-time framework for steering diffusion models with reward functions. FK steering works by sampling a system of multiple interacting diffusion processes, called particles, and resampling particles at intermediate steps based on scores computed using functions called potentials. Potentials are defined using rewards for intermediate states and are selected such that a high value indicates that the particle will yield a high-reward sample. We explore various choices of potentials, intermediate rewards, and samplers. We evaluate FK steering on text-to-image and text diffusion models. For steering text-to-image models with a human preference reward, we find that FK steering a 0.8B parameter model outperforms a 2.6B parameter fine-tuned model on prompt fidelity, with faster sampling and no training. For steering text diffusion models with rewards for text quality and specific text attributes, we find that FK steering generates lower perplexity, more linguistically acceptable outputs and enables gradient-free control of attributes like toxicity. Our results demonstrate that inference-time scaling and steering of diffusion models, even with off-the-shelf rewards, can provide significant sample quality gains and controllability benefits. Code is available at https://github.com/zacharyhorvitz/Fk-Diffusion-Steering .
♻ ☆ Consistency of Responses and Continuations Generated by Large Language Models on Social Media
Large Language Models (LLMs) demonstrate remarkable capabilities in text generation, yet their emotional consistency and semantic coherence in social media contexts remain insufficiently understood. This study investigates how LLMs handle emotional content and maintain semantic relationships through continuation and response tasks using two open-source models: Gemma and Llama. By analyzing climate change discussions from Twitter and Reddit, we examine emotional transitions, intensity patterns, and semantic similarity between human-authored and LLM-generated content. Our findings reveal that while both models maintain high semantic coherence, they exhibit distinct emotional patterns: Gemma shows a tendency toward negative emotion amplification, particularly anger, while maintaining certain positive emotions like optimism. Llama demonstrates superior emotional preservation across a broader spectrum of affects. Both models systematically generate responses with attenuated emotional intensity compared to human-authored content and show a bias toward positive emotions in response tasks. Additionally, both models maintain strong semantic similarity with original texts, though performance varies between continuation and response tasks. These findings provide insights into LLMs' emotional and semantic processing capabilities, with implications for their deployment in social media contexts and human-AI interaction design.
♻ ☆ MSA-ASR: Efficient Multilingual Speaker Attribution with frozen ASR Models ICASSP 2025
Speaker-attributed automatic speech recognition (SA-ASR) aims to transcribe speech while assigning transcripts to the corresponding speakers accurately. Existing methods often rely on complex modular systems or require extensive fine-tuning of joint modules, limiting their adaptability and general efficiency. This paper introduces a novel approach, leveraging a frozen multilingual ASR model to incorporate speaker attribution into the transcriptions, using only standard monolingual ASR datasets. Our method involves training a speaker module to predict speaker embeddings based on weak labels without requiring additional ASR model modifications. Despite being trained exclusively with non-overlapping monolingual data, our approach effectively extracts speaker attributes across diverse multilingual datasets, including those with overlapping speech. Experimental results demonstrate competitive performance compared to strong baselines, highlighting the model's robustness and potential for practical applications.
comment: Accepted at ICASSP 2025
♻ ☆ Beyond Boundaries: Learning a Universal Entity Taxonomy across Datasets and Languages for Open Named Entity Recognition COLING 2025
Open Named Entity Recognition (NER), which involves identifying arbitrary types of entities from arbitrary domains, remains challenging for Large Language Models (LLMs). Recent studies suggest that fine-tuning LLMs on extensive NER data can boost their performance. However, training directly on existing datasets neglects their inconsistent entity definitions and redundant data, limiting LLMs to dataset-specific learning and hindering out-of-domain adaptation. To address this, we present B2NERD, a compact dataset designed to guide LLMs' generalization in Open NER under a universal entity taxonomy. B2NERD is refined from 54 existing English and Chinese datasets using a two-step process. First, we detect inconsistent entity definitions across datasets and clarify them by distinguishable label names to construct a universal taxonomy of 400+ entity types. Second, we address redundancy using a data pruning strategy that selects fewer samples with greater category and semantic diversity. Comprehensive evaluation shows that B2NERD significantly enhances LLMs' Open NER capabilities. Our B2NER models, trained on B2NERD, outperform GPT-4 by 6.8-12.0 F1 points and surpass previous methods in 3 out-of-domain benchmarks across 15 datasets and 6 languages. The data, models, and code are publicly available at https://github.com/UmeanNever/B2NER.
comment: Accepted at COLING 2025. Camera-ready version updated. Project page: https://github.com/UmeanNever/B2NER
♻ ☆ IndoNLP 2025: Shared Task on Real-Time Reverse Transliteration for Romanized Indo-Aryan languages
The paper overviews the shared task on Real-Time Reverse Transliteration for Romanized Indo-Aryan languages. It focuses on the reverse transliteration of low-resourced languages in the Indo-Aryan family to their native scripts. Typing Romanized Indo-Aryan languages using ad-hoc transliterals and achieving accurate native scripts are complex and often inaccurate processes with the current keyboard systems. This task aims to introduce and evaluate a real-time reverse transliterator that converts Romanized Indo-Aryan languages to their native scripts, improving the typing experience for users. Out of 11 registered teams, four teams participated in the final evaluation phase with transliteration models for Sinhala, Hindi and Malayalam. These proposed solutions not only solve the issue of ad-hoc transliteration but also empower low-resource language usability in the digital arena.
comment: 7 Pages, 1 Figure, 3 Tables
♻ ☆ Mind the Error! Detection and Localization of Instruction Errors in Vision-and-Language Navigation IROS'24
Vision-and-Language Navigation in Continuous Environments (VLN-CE) is one of the most intuitive yet challenging embodied AI tasks. Agents are tasked to navigate towards a target goal by executing a set of low-level actions, following a series of natural language instructions. All VLN-CE methods in the literature assume that language instructions are exact. However, in practice, instructions given by humans can contain errors when describing a spatial environment due to inaccurate memory or confusion. Current VLN-CE benchmarks do not address this scenario, making the state-of-the-art methods in VLN-CE fragile in the presence of erroneous instructions from human users. For the first time, we propose a novel benchmark dataset that introduces various types of instruction errors considering potential human causes. This benchmark provides valuable insight into the robustness of VLN systems in continuous environments. We observe a noticeable performance drop (up to -25%) in Success Rate when evaluating the state-of-the-art VLN-CE methods on our benchmark. Moreover, we formally define the task of Instruction Error Detection and Localization, and establish an evaluation protocol on top of our benchmark dataset. We also propose an effective method, based on a cross-modal transformer architecture, that achieves the best performance in error detection and localization, compared to baselines. Surprisingly, our proposed method has revealed errors in the validation set of the two commonly used datasets for VLN-CE, i.e., R2R-CE and RxR-CE, demonstrating the utility of our technique in other tasks. Code and dataset available at https://intelligolabs.github.io/R2RIE-CE
comment: 3 figures, 8 pages. Accepted at IROS'24
♻ ☆ How to Build an AI Tutor That Can Adapt to Any Course Using Knowledge Graph-Enhanced Retrieval-Augmented Generation (KG-RAG)
This paper introduces a novel framework for adaptable AI tutors using Knowledge Graph-enhanced Retrieval-Augmented Generation (KG-RAG). This approach addresses the critical challenges of information hallucination and limited course-specific adaptation prevalent in Large Language Model (LLM)-based tutoring systems. By integrating Knowledge Graphs (KGs) with RAG, we provide a structured representation of course concepts and their interrelationships, grounding the AI tutor's responses in relevant, validated material. We leverage Qwen2.5, a powerful and cost-effective LLM, within our KG-RAG framework. A user study (n=50) demonstrated positive student feedback regarding answer relevance, ease of use, and overall satisfaction. This KG-RAG framework offers a promising pathway towards personalized learning experiences and broader access to high-quality education.
comment: 6 pages, 5 figures
♻ ☆ Parallelizing Linear Transformers with the Delta Rule over Sequence Length
Transformers with linear attention (i.e., linear transformers) and state-space models have recently been suggested as a viable linear-time alternative to transformers with softmax attention. However, these models still underperform transformers especially on tasks that require in-context retrieval. While more expressive variants of linear transformers which replace the additive update in linear transformers with the delta rule (DeltaNet) have been found to be more effective at associative recall, existing algorithms for training such models do not parallelize over sequence length and are thus inefficient to train on modern hardware. This work describes a hardware-efficient algorithm for training linear transformers with the delta rule, which exploits a memory-efficient representation for computing products of Householder matrices. This algorithm allows us to scale up DeltaNet to standard language modeling settings. We train a 1.3B model for 100B tokens and find that it outperforms recent linear-time baselines such as Mamba and GLA in terms of perplexity and zero-shot performance on downstream tasks. We also experiment with two hybrid models which combine DeltaNet layers with (1) sliding-window attention layers every other layer or (2) two global attention layers, and find that these hybrids outperform strong transformer baselines.
comment: Final camera ready
♻ ☆ Toward Automated Simulation Research Workflow through LLM Prompt Engineering Design
The advent of Large Language Models (LLMs) has created new opportunities for the automation of scientific research spanning both experimental processes and computational simulations. This study explores the feasibility of constructing an autonomous simulation agent (ASA) powered by LLMs through prompt engineering and automated program design to automate the entire simulation research process according to a human-provided research plan. This process includes experimental design, remote upload and simulation execution, data analysis, and report compilation. Using a well-studied simulation problem of polymer chain conformations as a test case, we assessed the long-task completion and reliability of ASAs powered by different LLMs, including GPT-4o, Claude-3.5, etc. Our findings revealed that ASA-GPT-4o achieved near-flawless execution on designated research missions, underscoring the potential of methods like ASA to achieve automation in simulation research processes to enhance research efficiency. The outlined automation can be iteratively performed for up to 20 cycles without human intervention, illustrating the potential of ASA for long-task workflow automation. Additionally, we discussed the intrinsic traits of ASA in managing extensive tasks, focusing on self-validation mechanisms, and the balance between local attention and global oversight.
comment: The source code and example results of ASA can be found at https://github.com/zokaraa/autonomous_simulation_agent
♻ ☆ AfriHate: A Multilingual Collection of Hate Speech and Abusive Language Datasets for African Languages
Hate speech and abusive language are global phenomena that need socio-cultural background knowledge to be understood, identified, and moderated. However, in many regions of the Global South, there have been several documented occurrences of (1) absence of moderation and (2) censorship due to the reliance on keyword spotting out of context. Further, high-profile individuals have frequently been at the center of the moderation process, while large and targeted hate speech campaigns against minorities have been overlooked. These limitations are mainly due to the lack of high-quality data in the local languages and the failure to include local communities in the collection, annotation, and moderation processes. To address this issue, we present AfriHate: a multilingual collection of hate speech and abusive language datasets in 15 African languages. Each instance in AfriHate is annotated by native speakers familiar with the local culture. We report the challenges related to the construction of the datasets and present various classification baseline results with and without using LLMs. The datasets, individual annotations, and hate speech and offensive language lexicons are available on https://github.com/AfriHate/AfriHate
♻ ☆ SelectIT: Selective Instruction Tuning for LLMs via Uncertainty-Aware Self-Reflection NeurIPS 2024
Instruction tuning (IT) is crucial to tailoring large language models (LLMs) towards human-centric interactions. Recent advancements have shown that the careful selection of a small, high-quality subset of IT data can significantly enhance the performance of LLMs. Despite this, common approaches often rely on additional models or data, which increases costs and limits widespread adoption. In this work, we propose a novel approach, termed SelectIT, that capitalizes on the foundational capabilities of the LLM itself. Specifically, we exploit the intrinsic uncertainty present in LLMs to more effectively select high-quality IT data, without the need for extra resources. Furthermore, we introduce a curated IT dataset, the Selective Alpaca, created by applying SelectIT to the Alpaca-GPT4 dataset. Empirical results demonstrate that IT using Selective Alpaca leads to substantial model ability enhancement. The robustness of SelectIT has also been corroborated in various foundation models and domain-specific tasks. Our findings suggest that longer and more computationally intensive IT data may serve as superior sources of IT, offering valuable insights for future research in this area. Data, code, and scripts are freely available at https://github.com/Blue-Raincoat/SelectIT.
comment: Accepted to NeurIPS 2024
♻ ☆ Mitigating Knowledge Conflicts in Language Model-Driven Question Answering
In the context of knowledge-driven seq-to-seq generation tasks, such as document-based question answering and document summarization systems, two fundamental knowledge sources play crucial roles: the inherent knowledge embedded within model parameters and the external knowledge obtained through context. Recent studies revealed a significant challenge: when there exists a misalignment between the model's inherent knowledge and the ground truth answers in training data, the system may exhibit problematic behaviors during inference, such as ignoring input context, or generating unfaithful content. Our investigation proposes a strategy to minimize hallucination by building explicit connection between source inputs and generated outputs. We specifically target a common hallucination pattern in question answering, examining how the correspondence between entities and their contexts during model training influences the system's performance at inference time.
comment: revised version, more figures
♻ ☆ TANQ: An open domain dataset of table answered questions
Language models, potentially augmented with tool usage such as retrieval are becoming the go-to means of answering questions. Understanding and answering questions in real-world settings often requires retrieving information from different sources, processing and aggregating data to extract insights, and presenting complex findings in form of structured artifacts such as novel tables, charts, or infographics. In this paper, we introduce TANQ, the first open domain question answering dataset where the answers require building tables from information across multiple sources. We release the full source attribution for every cell in the resulting table and benchmark state-of-the-art language models in open, oracle, and closed book setups. Our best-performing baseline, GPT4 reaches an overall F1 score of 29.1, lagging behind human performance by 19.7 points. We analyse baselines' performance across different dataset attributes such as different skills required for this task, including multi-hop reasoning, math operations, and unit conversions. We further discuss common failures in model-generated answers, suggesting that TANQ is a complex task with many challenges ahead.
comment: 10 pages
♻ ☆ Salmon: A Suite for Acoustic Language Model Evaluation ICASSP 2025
Speech language models have recently demonstrated great potential as universal speech processing systems. Such models have the ability to model the rich acoustic information existing in audio signals, beyond spoken content, such as emotion, background noise, etc. Despite this, evaluation benchmarks which evaluate awareness to a wide range of acoustic aspects, are lacking. To help bridge this gap, we introduce SALMon, a novel evaluation suite encompassing background noise, emotion, speaker identity and room impulse response. The proposed benchmarks both evaluate the consistency of the inspected element and how much it matches the spoken text. We follow a modelling based approach, measuring whether a model gives correct samples higher scores than incorrect ones. This approach makes the benchmark fast to compute even for large models. We evaluated several speech language models on SALMon, thus highlighting the strengths and weaknesses of each evaluated method. We make the code and data publicly available at https://pages.cs.huji.ac.il/adiyoss-lab/salmon/ .
comment: ICASSP 2025, project page - https://pages.cs.huji.ac.il/adiyoss-lab/salmon/
♻ ☆ Noise-powered Multi-modal Knowledge Graph Representation Framework COLING 2025
The rise of Multi-modal Pre-training highlights the necessity for a unified Multi-Modal Knowledge Graph (MMKG) representation learning framework. Such a framework is essential for embedding structured knowledge into multi-modal Large Language Models effectively, alleviating issues like knowledge misconceptions and multi-modal hallucinations. In this work, we explore the efficacy of models in accurately embedding entities within MMKGs through two pivotal tasks: Multi-modal Knowledge Graph Completion (MKGC) and Multi-modal Entity Alignment (MMEA). Building on this foundation, we propose a novel SNAG method that utilizes a Transformer-based architecture equipped with modality-level noise masking to robustly integrate multi-modal entity features in KGs. By incorporating specific training objectives for both MKGC and MMEA, our approach achieves SOTA performance across a total of ten datasets, demonstrating its versatility. Moreover, SNAG can not only function as a standalone model but also enhance other existing methods, providing stable performance improvements. Code and data are available at https://github.com/zjukg/SNAG.
comment: COLING 2025 Accepted, Repo is available at https://github.com/zjukg/SNAG
♻ ☆ Comparative Analysis of Listwise Reranking with Large Language Models in Limited-Resource Language Contexts
Large Language Models (LLMs) have demonstrated significant effectiveness across various NLP tasks, including text ranking. This study assesses the performance of large language models (LLMs) in listwise reranking for limited-resource African languages. We compare proprietary models RankGPT3.5, Rank4o-mini, RankGPTo1-mini and RankClaude-sonnet in cross-lingual contexts. Results indicate that these LLMs significantly outperform traditional baseline methods such as BM25-DT in most evaluation metrics, particularly in nDCG@10 and MRR@100. These findings highlight the potential of LLMs in enhancing reranking tasks for low-resource languages and offer insights into cost-effective solutions.
♻ ☆ Do Large Language Models Mirror Cognitive Language Processing?
Large Language Models (LLMs) have demonstrated remarkable abilities in text comprehension and logical reasoning, indicating that the text representations learned by LLMs can facilitate their language processing capabilities. In neuroscience, brain cognitive processing signals are typically utilized to study human language processing. Therefore, it is natural to ask how well the text embeddings from LLMs align with the brain cognitive processing signals, and how training strategies affect the LLM-brain alignment? In this paper, we employ Representational Similarity Analysis (RSA) to measure the alignment between 23 mainstream LLMs and fMRI signals of the brain to evaluate how effectively LLMs simulate cognitive language processing. We empirically investigate the impact of various factors (e.g., pre-training data size, model scaling, alignment training, and prompts) on such LLM-brain alignment. Experimental results indicate that pre-training data size and model scaling are positively correlated with LLM-brain similarity, and alignment training can significantly improve LLM-brain similarity. Explicit prompts contribute to the consistency of LLMs with brain cognitive language processing, while nonsensical noisy prompts may attenuate such alignment. Additionally, the performance of a wide range of LLM evaluations (e.g., MMLU, Chatbot Arena) is highly correlated with the LLM-brain similarity.
♻ ☆ Natural Language Outlines for Code: Literate Programming in the LLM Era
We propose using natural language outlines as a novel modality and interaction surface for providing AI assistance to developers throughout the software development process. An NL outline for a code function comprises multiple statements written in concise prose, which partition the code and summarize its main ideas in the style of literate programming. Crucially, we find that modern LLMs can generate accurate and high-quality NL outlines in practice. Moreover, NL outlines enable a bidirectional sync between code and NL, allowing changes in one to be automatically reflected in the other. We discuss many use cases for NL outlines: they can accelerate understanding and navigation of code and diffs, simplify code maintenance, augment code search, steer code generation, and more. We then propose and compare multiple LLM prompting techniques for generating outlines and ask professional developers to judge outline quality. Finally, we present two case studies applying NL outlines toward code review and malware detection.
♻ ☆ Counterfactual Debating with Preset Stances for Hallucination Elimination of LLMs COLING 2025
Large Language Models (LLMs) excel in various natural language processing tasks but struggle with hallucination issues. Existing solutions have considered utilizing LLMs' inherent reasoning abilities to alleviate hallucination, such as self-correction and diverse sampling methods. However, these methods often overtrust LLMs' initial answers due to inherent biases. The key to alleviating this issue lies in overriding LLMs' inherent biases for answer inspection. To this end, we propose a CounterFactual Multi-Agent Debate (CFMAD) framework. CFMAD presets the stances of LLMs to override their inherent biases by compelling LLMs to generate justifications for a predetermined answer's correctness. The LLMs with different predetermined stances are engaged with a skeptical critic for counterfactual debate on the rationality of generated justifications. Finally, the debate process is evaluated by a third-party judge to determine the final answer. Extensive experiments on four datasets of three tasks demonstrate the superiority of CFMAD over existing methods.
comment: accepted by COLING 2025
♻ ☆ KaLM-Embedding: Superior Training Data Brings A Stronger Embedding Model
As retrieval-augmented generation prevails in large language models, embedding models are becoming increasingly crucial. Despite the growing number of general embedding models, prior work often overlooks the critical role of training data quality. In this work, we introduce KaLM-Embedding, a general multilingual embedding model that leverages a large quantity of cleaner, more diverse, and domain-specific training data. Our model has been trained with key techniques proven to enhance performance: (1) persona-based synthetic data to create diversified examples distilled from LLMs, (2) ranking consistency filtering to remove less informative samples, and (3) semi-homogeneous task batch sampling to improve training efficacy. Departing from traditional BERT-like architectures, we adopt Qwen2-0.5B as the pre-trained model, facilitating the adaptation of auto-regressive language models for general embedding tasks. Extensive evaluations of the MTEB benchmark across multiple languages show that our model outperforms others of comparable size, setting a new standard for multilingual embedding models with <1B parameters.
comment: Technical Report. 23 pages, 6 figures, 10 tables
♻ ☆ A Multi-Modal AI Copilot for Single-Cell Analysis with Instruction Following
Large language models excel at interpreting complex natural language instructions, enabling them to perform a wide range of tasks. In the life sciences, single-cell RNA sequencing (scRNA-seq) data serves as the "language of cellular biology", capturing intricate gene expression patterns at the single-cell level. However, interacting with this "language" through conventional tools is often inefficient and unintuitive, posing challenges for researchers. To address these limitations, we present InstructCell, a multi-modal AI copilot that leverages natural language as a medium for more direct and flexible single-cell analysis. We construct a comprehensive multi-modal instruction dataset that pairs text-based instructions with scRNA-seq profiles from diverse tissues and species. Building on this, we develop a multi-modal cell language architecture capable of simultaneously interpreting and processing both modalities. InstructCell empowers researchers to accomplish critical tasks-such as cell type annotation, conditional pseudo-cell generation, and drug sensitivity prediction-using straightforward natural language commands. Extensive evaluations demonstrate that InstructCell consistently meets or exceeds the performance of existing single-cell foundation models, while adapting to diverse experimental conditions. More importantly, InstructCell provides an accessible and intuitive tool for exploring complex single-cell data, lowering technical barriers and enabling deeper biological insights.
comment: 37 pages; 13 figures; Code: https://github.com/zjunlp/Instructcell, Models: https://huggingface.co/zjunlp/Instructcell-chat, https://huggingface.co/zjunlp/InstructCell-instruct
♻ ☆ Understanding Emergent Abilities of Language Models from the Loss Perspective NeurIPS 2024
Recent studies have put into question the belief that emergent abilities in language models are exclusive to large models. This skepticism arises from two observations: 1) smaller models can also exhibit high performance on emergent abilities and 2) there is doubt on the discontinuous metrics used to measure these abilities. In this paper, we propose to study emergent abilities in the lens of pre-training loss, instead of model size or training compute. We demonstrate that the Transformer models with the same pre-training loss, but different model and data sizes, generate the same performance on various downstream tasks, with a fixed data corpus, tokenization, and model architecture. We also discover that a model exhibits emergent abilities on certain tasks -- regardless of the continuity of metrics -- when its pre-training loss falls below a specific threshold. Before reaching this threshold, its performance remains at the level of random guessing. This inspires us to redefine emergent abilities as those that manifest in models with lower pre-training losses, highlighting that these abilities cannot be predicted by merely extrapolating the performance trends of models with higher pre-training losses.
comment: 23 pages, 8 figures. Accepted in NeurIPS 2024
♻ ☆ Con-ReCall: Detecting Pre-training Data in LLMs via Contrastive Decoding
The training data in large language models is key to their success, but it also presents privacy and security risks, as it may contain sensitive information. Detecting pre-training data is crucial for mitigating these concerns. Existing methods typically analyze target text in isolation or solely with non-member contexts, overlooking potential insights from simultaneously considering both member and non-member contexts. While previous work suggested that member contexts provide little information due to the minor distributional shift they induce, our analysis reveals that these subtle shifts can be effectively leveraged when contrasted with non-member contexts. In this paper, we propose Con-ReCall, a novel approach that leverages the asymmetric distributional shifts induced by member and non-member contexts through contrastive decoding, amplifying subtle differences to enhance membership inference. Extensive empirical evaluations demonstrate that Con-ReCall achieves state-of-the-art performance on the WikiMIA benchmark and is robust against various text manipulation techniques.
♻ ☆ Compositional Automata Embeddings for Goal-Conditioned Reinforcement Learning
Goal-conditioned reinforcement learning is a powerful way to control an AI agent's behavior at runtime. That said, popular goal representations, e.g., target states or natural language, are either limited to Markovian tasks or rely on ambiguous task semantics. We propose representing temporal goals using compositions of deterministic finite automata (cDFAs) and use cDFAs to guide RL agents. cDFAs balance the need for formal temporal semantics with ease of interpretation: if one can understand a flow chart, one can understand a cDFA. On the other hand, cDFAs form a countably infinite concept class with Boolean semantics, and subtle changes to the automaton can result in very different tasks, making them difficult to condition agent behavior on. To address this, we observe that all paths through a DFA correspond to a series of reach-avoid tasks and propose pre-training graph neural network embeddings on "reach-avoid derived" DFAs. Through empirical evaluation, we demonstrate that the proposed pre-training method enables zero-shot generalization to various cDFA task classes and accelerated policy specialization without the myopic suboptimality of hierarchical methods.
♻ ☆ 2 OLMo 2 Furious
We present OLMo 2, the next generation of our fully open language models. OLMo 2 includes dense autoregressive models with improved architecture and training recipe, pretraining data mixtures, and instruction tuning recipes. Our modified model architecture and training recipe achieve both better training stability and improved per-token efficiency. Our updated pretraining data mixture introduces a new, specialized data mix called Dolmino Mix 1124, which significantly improves model capabilities across many downstream task benchmarks when introduced via late-stage curriculum training (i.e. specialized data during the annealing phase of pretraining). Finally, we incorporate best practices from T\"ulu 3 to develop OLMo 2-Instruct, focusing on permissive data and extending our final-stage reinforcement learning with verifiable rewards (RLVR). Our OLMo 2 base models sit at the Pareto frontier of performance to compute, often matching or outperforming open-weight only models like Llama 3.1 and Qwen 2.5 while using fewer FLOPs and with fully transparent training data, code, and recipe. Our fully open OLMo 2-Instruct models are competitive with or surpassing open-weight only models of comparable size, including Qwen 2.5, Llama 3.1 and Gemma 2. We release all OLMo 2 artifacts openly -- models at 7B and 13B scales, both pretrained and post-trained, including their full training data, training code and recipes, training logs and thousands of intermediate checkpoints. The final instruction model is available on the Ai2 Playground as a free research demo.
comment: Model demo available at playground.allenai.org
♻ ☆ Automated Review Generation Method Based on Large Language Models
Literature research, vital for scientific work, faces the challenge of surging information volumes exceeding researchers' processing capabilities. We present an automated review generation method based on large language models (LLMs) to overcome efficiency bottlenecks and reduce cognitive load. Our statistically validated evaluation framework demonstrates that the generated reviews match or exceed manual quality, offering broad applicability across research fields without requiring users' domain knowledge. Applied to propane dehydrogenation (PDH) catalysts, our method swiftly analyzed 343 articles, averaging seconds per article per LLM account, producing comprehensive reviews spanning 35 topics, with extended analysis of 1041 articles providing insights into catalysts' properties. Through multi-layered quality control, we effectively mitigated LLMs' hallucinations, with expert verification confirming accuracy and citation integrity while demonstrating hallucination risks reduced to below 0.5\% with 95\% confidence. Released Windows application enables one-click review generation, enhancing research productivity and literature recommendation efficiency while setting the stage for broader scientific explorations.
comment: 21 pages, 5 figures, 1 tables Code: https://github.com/TJU-ECAT-AI/AutomaticReviewGeneration Data: https://github.com/TJU-ECAT-AI/AutomaticReviewGenerationData This research has been invited for a Short Oral presentation at the 18th ICC - International Congress on Catalysis, taking place in Lyon, France from July 14-19, 2024
♻ ☆ Unmasking the Imposters: How Censorship and Domain Adaptation Affect the Detection of Machine-Generated Tweets
The rapid development of large language models (LLMs) has significantly improved the generation of fluent and convincing text, raising concerns about their potential misuse on social media platforms. We present a comprehensive methodology for creating nine Twitter datasets to examine the generative capabilities of four prominent LLMs: Llama 3, Mistral, Qwen2, and GPT4o. These datasets encompass four censored and five uncensored model configurations, including 7B and 8B parameter base-instruction models of the three open-source LLMs. Additionally, we perform a data quality analysis to assess the characteristics of textual outputs from human, "censored," and "uncensored" models, employing semantic meaning, lexical richness, structural patterns, content characteristics, and detector performance metrics to identify differences and similarities. Our evaluation demonstrates that "uncensored" models significantly undermine the effectiveness of automated detection methods. This study addresses a critical gap by exploring smaller open-source models and the ramifications of "uncensoring," providing valuable insights into how domain adaptation and content moderation strategies influence both the detectability and structural characteristics of machine-generated text.
♻ ☆ PASS: Presentation Automation for Slide Generation and Speech
In today's fast-paced world, effective presentations have become an essential tool for communication in both online and offline meetings. The crafting of a compelling presentation requires significant time and effort, from gathering key insights to designing slides that convey information clearly and concisely. However, despite the wealth of resources available, people often find themselves manually extracting crucial points, analyzing data, and organizing content in a way that ensures clarity and impact. Furthermore, a successful presentation goes beyond just the slides; it demands rehearsal and the ability to weave a captivating narrative to fully engage the audience. Although there has been some exploration of automating document-to-slide generation, existing research is largely centered on converting research papers. In addition, automation of the delivery of these presentations has yet to be addressed. We introduce PASS, a pipeline used to generate slides from general Word documents, going beyond just research papers, which also automates the oral delivery of the generated slides. PASS analyzes user documents to create a dynamic, engaging presentation with an AI-generated voice. Additionally, we developed an LLM-based evaluation metric to assess our pipeline across three critical dimensions of presentations: relevance, coherence, and redundancy. The data and codes are available at https://github.com/AggarwalTushar/PASS.
♻ ☆ Contextual Evaluation of Large Language Models for Classifying Tropical and Infectious Diseases NeurIPS 2024
While large language models (LLMs) have shown promise for medical question answering, there is limited work focused on tropical and infectious disease-specific exploration. We build on an opensource tropical and infectious diseases (TRINDs) dataset, expanding it to include demographic and semantic clinical and consumer augmentations yielding 11000+ prompts. We evaluate LLM performance on these, comparing generalist and medical LLMs, as well as LLM outcomes to human experts. We demonstrate through systematic experimentation, the benefit of contextual information such as demographics, location, gender, risk factors for optimal LLM response. Finally we develop a prototype of TRINDs-LM, a research tool that provides a playground to navigate how context impacts LLM outputs for health.
comment: Accepted at 2 NeurIPS 2024 workshops: Generative AI for Health Workshop and Workshop on Advancements In Medical Foundation Models: Explainability, Robustness, Security, and Beyond
Machine Learning 155
☆ Towards Fast, Specialized Machine Learning Force Fields: Distilling Foundation Models via Energy Hessians ICLR 2025
The foundation model (FM) paradigm is transforming Machine Learning Force Fields (MLFFs), leveraging general-purpose representations and scalable training to perform a variety of computational chemistry tasks. Although MLFF FMs have begun to close the accuracy gap relative to first-principles methods, there is still a strong need for faster inference speed. Additionally, while research is increasingly focused on general-purpose models which transfer across chemical space, practitioners typically only study a small subset of systems at a given time. This underscores the need for fast, specialized MLFFs relevant to specific downstream applications, which preserve test-time physical soundness while maintaining train-time scalability. In this work, we introduce a method for transferring general-purpose representations from MLFF foundation models to smaller, faster MLFFs specialized to specific regions of chemical space. We formulate our approach as a knowledge distillation procedure, where the smaller "student" MLFF is trained to match the Hessians of the energy predictions of the "teacher" foundation model. Our specialized MLFFs can be up to 20 $\times$ faster than the original foundation model, while retaining, and in some cases exceeding, its performance and that of undistilled models. We also show that distilling from a teacher model with a direct force parameterization into a student model trained with conservative forces (i.e., computed as derivatives of the potential energy) successfully leverages the representations from the large-scale teacher for improved accuracy, while maintaining energy conservation during test-time molecular dynamics simulations. More broadly, our work suggests a new paradigm for MLFF development, in which foundation models are released along with smaller, specialized simulation "engines" for common chemical subsets.
comment: Under Review at ICLR 2025
☆ Improving Stability Estimates in Adversarial Explainable AI through Alternate Search Methods
Advances in the effectiveness of machine learning models have come at the cost of enormous complexity resulting in a poor understanding of how they function. Local surrogate methods have been used to approximate the workings of these complex models, but recent work has revealed their vulnerability to adversarial attacks where the explanation produced is appreciably different while the meaning and structure of the complex model's output remains similar. This prior work has focused on the existence of these weaknesses but not on their magnitude. Here we explore using an alternate search method with the goal of finding minimum viable perturbations, the fewest perturbations necessary to achieve a fixed similarity value between the original and altered text's explanation. Intuitively, a method that requires fewer perturbations to expose a given level of instability is inferior to one which requires more. This nuance allows for superior comparisons of the stability of explainability methods.
comment: 9 pages, 3 figures, 5 tables. arXiv admin note: text overlap with arXiv:2406.15839
☆ CrystalGRW: Generative Modeling of Crystal Structures with Targeted Properties via Geodesic Random Walks
Determining whether a candidate crystalline material is thermodynamically stable depends on identifying its true ground-state structure, a central challenge in computational materials science. We introduce CrystalGRW, a diffusion-based generative model on Riemannian manifolds that proposes novel crystal configurations and can predict stable phases validated by density functional theory. The crystal properties, such as fractional coordinates, atomic types, and lattice matrices, are represented on suitable Riemannian manifolds, ensuring that new predictions generated through the diffusion process preserve the periodicity of crystal structures. We incorporate an equivariant graph neural network to also account for rotational and translational symmetries during the generation process. CrystalGRW demonstrates the ability to generate realistic crystal structures that are close to their ground states with accuracy comparable to existing models, while also enabling conditional control, such as specifying a desired crystallographic point group. These features help accelerate materials discovery and inverse design by offering stable, symmetry-consistent crystal candidates for experimental validation.
comment: 10+12 pages, 10 figures
☆ VECT-GAN: A variationally encoded generative model for overcoming data scarcity in pharmaceutical science
Data scarcity in pharmaceutical research has led to reliance on labour-intensive trial and error approaches for development rather than data driven methods. While Machine Learning offers a solution, existing datasets are often small and noisy, limiting their utility. To address this, we developed a Variationally Encoded Conditional Tabular Generative Adversarial Network (VECT GAN), a novel generative model specifically designed for augmenting small, noisy datasets. We introduce a pipeline where data is augmented before regression model development and demonstrate that this consistently and significantly improves performance over other state of the art tabular generative models. We apply this pipeline across six pharmaceutical datasets, and highlight its real-world applicability by developing novel polymers with medically desirable mucoadhesive properties, which we made and experimentally characterised. Additionally, we pre-train the model on the ChEMBL database of drug-like molecules, leveraging knowledge distillation to enhance its generalisability, making it readily available for use on pharmaceutical datasets containing small molecules, which is an extremely common pharmaceutical task. We demonstrate the power of synthetic data for regularising small tabular datasets, highlighting its potential to become standard practice in pharmaceutical model development, and make our method, including VECT GAN pretrained on ChEMBL available as a pip package.
comment: 30 pages, 6 primary figures, 3 supplementary figures
☆ Trusted Machine Learning Models Unlock Private Inference for Problems Currently Infeasible with Cryptography
We often interact with untrusted parties. Prioritization of privacy can limit the effectiveness of these interactions, as achieving certain goals necessitates sharing private data. Traditionally, addressing this challenge has involved either seeking trusted intermediaries or constructing cryptographic protocols that restrict how much data is revealed, such as multi-party computations or zero-knowledge proofs. While significant advances have been made in scaling cryptographic approaches, they remain limited in terms of the size and complexity of applications they can be used for. In this paper, we argue that capable machine learning models can fulfill the role of a trusted third party, thus enabling secure computations for applications that were previously infeasible. In particular, we describe Trusted Capable Model Environments (TCMEs) as an alternative approach for scaling secure computation, where capable machine learning model(s) interact under input/output constraints, with explicit information flow control and explicit statelessness. This approach aims to achieve a balance between privacy and computational efficiency, enabling private inference where classical cryptographic solutions are currently infeasible. We describe a number of use cases that are enabled by TCME, and show that even some simple classic cryptographic problems can already be solved with TCME. Finally, we outline current limitations and discuss the path forward in implementing them.
☆ Training-Aware Risk Control for Intensity Modulated Radiation Therapies Quality Assurance with Conformal Prediction
Measurement quality assurance (QA) practices play a key role in the safe use of Intensity Modulated Radiation Therapies (IMRT) for cancer treatment. These practices have reduced measurement-based IMRT QA failure below 1%. However, these practices are time and labor intensive which can lead to delays in patient care. In this study, we examine how conformal prediction methodologies can be used to robustly triage plans. We propose a new training-aware conformal risk control method by combining the benefit of conformal risk control and conformal training. We incorporate the decision making thresholds based on the gamma passing rate, along with the risk functions used in clinical evaluation, into the design of the risk control framework. Our method achieves high sensitivity and specificity and significantly reduces the number of plans needing measurement without generating a huge confidence interval. Our results demonstrate the validity and applicability of conformal prediction methods for improving efficiency and reducing the workload of the IMRT QA process.
comment: 2024 Machine Learning for Health Symposium
☆ Kolmogorov-Arnold Networks for Time Series Granger Causality Inference
We introduce Granger Causality Kolmogorov-Arnold Networks (GCKAN), an innovative architecture that extends the recently proposed Kolmogorov-Arnold Networks (KAN) to the domain of causal inference. By extracting base weights from KAN layers and incorporating the sparsity-inducing penalty along with ridge regularization, GCKAN infers the Granger causality from time series while enabling automatic time lag selection. Additionally, we propose an algorithm leveraging time-reversed Granger causality to enhance inference accuracy. The algorithm compares prediction and sparse-inducing losses derived from the original and time-reversed series, automatically selecting the casual relationship with the higher score or integrating the results to mitigate spurious connectivities. Comprehensive experiments conducted on Lorenz-96, gene regulatory networks, fMRI BOLD signals, and VAR datasets demonstrate that the proposed model achieves competitive performance to state-of-the-art methods in inferring Granger causality from nonlinear, high-dimensional, and limited-sample time series.
☆ Computing Approximated Fixpoints via Dampened Mann Iteration
Fixpoints are ubiquitous in computer science and when dealing with quantitative semantics and verification one is commonly led to consider least fixpoints of (higher-dimensional) functions over the nonnegative reals. We show how to approximate the least fixpoint of such functions, focusing on the case in which they are not known precisely, but represented by a sequence of approximating functions that converge to them. We concentrate on monotone and non-expansive functions, for which uniqueness of fixpoints is not guaranteed and standard fixpoint iteration schemes might get stuck at a fixpoint that is not the least. Our main contribution is the identification of an iteration scheme, a variation of Mann iteration with a dampening factor, which, under suitable conditions, is shown to guarantee convergence to the least fixpoint of the function of interest. We then argue that these results are relevant in the context of model-based reinforcement learning for Markov decision processes (MDPs), showing that the proposed iteration scheme instantiates to MDPs and allows us to derive convergence to the optimal expected return. More generally, we show that our results can be used to iterate to the least fixpoint almost surely for systems where the function of interest can be approximated with given probabilistic error bounds, as it happens for probabilistic systems, such as simple stochastic games, that can be explored via sampling.
☆ A Reinforcement Learning Approach to Quiet and Safe UAM Traffic Management
Urban air mobility (UAM) is a transformative system that operates various small aerial vehicles in urban environments to reshape urban transportation. However, integrating UAM into existing urban environments presents a variety of complex challenges. Recent analyses of UAM's operational constraints highlight aircraft noise and system safety as key hurdles to UAM system implementation. Future UAM air traffic management schemes must ensure that the system is both quiet and safe. We propose a multi-agent reinforcement learning approach to manage UAM traffic, aiming at both vertical separation assurance and noise mitigation. Through extensive training, the reinforcement learning agent learns to balance the two primary objectives by employing altitude adjustments in a multi-layer UAM network. The results reveal the tradeoffs among noise impact, traffic congestion, and separation. Overall, our findings demonstrate the potential of reinforcement learning in mitigating UAM's noise impact while maintaining safe separation using altitude adjustments
comment: Paper presented at SciTech 2025
☆ Disentangling Exploration of Large Language Models by Optimal Exploitation
Exploration is a crucial skill for self-improvement and open-ended problem-solving. However, it remains uncertain whether large language models can effectively explore the state-space. Existing evaluations predominantly focus on the trade-off between exploration and exploitation, often assessed in multi-armed bandit problems. In contrast, this work isolates exploration as the sole objective, tasking the agent with delivering information that enhances future returns. For the evaluation, we propose to decompose missing rewards into exploration and exploitation components by measuring the optimal achievable return for the states already explored. Our experiments with various LLMs reveal that most models struggle to sufficiently explore the state-space and that weak exploration is insufficient. We observe a positive correlation between model size and exploration performance, with larger models demonstrating superior capabilities. Furthermore, we show that our decomposition provides insights into differences in behaviors driven by agent instructions during prompt engineering, offering a valuable tool for refining LLM performance in exploratory tasks.
☆ Modeling Melt Pool Features and Spatter Using Symbolic Regression and Machine Learning
Additive manufacturing (AM) is a rapidly evolving technology that has attracted applications across a wide range of fields due to its ability to fabricate complex geometries. However, one of the key challenges in AM is achieving consistent print quality. This inconsistency is often attributed to uncontrolled melt pool dynamics, partly caused by spatter which can lead to defects. Therefore, capturing and controlling the evolution of the melt pool is crucial for enhancing process stability and part quality. In this study, we developed a framework to support decision-making in AM operations, facilitating quality control and minimizing defects via machine learning (ML) and polynomial symbolic regression models. We implemented experimentally validated computational tools as a cost-effective approach to collect large datasets from laser powder bed fusion (LPBF) processes. For a dataset consisting of 281 process conditions, parameters such as melt pool dimensions (length, width, depth), melt pool geometry (area, volume), and volume indicated as spatter were extracted. Using machine learning (ML) and polynomial symbolic regression models, a high R2 of over 95 % was achieved in predicting the melt pool dimensions and geometry features for both the training and testing datasets, with either process conditions (power and velocity) or melt pool dimensions as the model inputs. In the case of volume indicated as spatter, R2 improved after logarithmic transforming the model inputs, which was either the process conditions or the melt pool dimensions. Among the investigated ML models, the ExtraTree model achieved the highest R2 values of 96.7 % and 87.5 %.
☆ GenAI Content Detection Task 3: Cross-Domain Machine-Generated Text Detection Challenge COLING 2025
Recently there have been many shared tasks targeting the detection of generated text from Large Language Models (LLMs). However, these shared tasks tend to focus either on cases where text is limited to one particular domain or cases where text can be from many domains, some of which may not be seen during test time. In this shared task, using the newly released RAID benchmark, we aim to answer whether or not models can detect generated text from a large, yet fixed, number of domains and LLMs, all of which are seen during training. Over the course of three months, our task was attempted by 9 teams with 23 detector submissions. We find that multiple participants were able to obtain accuracies of over 99% on machine-generated text from RAID while maintaining a 5% False Positive Rate -- suggesting that detectors are able to robustly detect text from many domains and models simultaneously. We discuss potential interpretations of this result and provide directions for future research.
comment: COLING 2025
☆ Projection Implicit Q-Learning with Support Constraint for Offline Reinforcement Learning
Offline Reinforcement Learning (RL) faces a critical challenge of extrapolation errors caused by out-of-distribution (OOD) actions. Implicit Q-Learning (IQL) algorithm employs expectile regression to achieve in-sample learning, effectively mitigating the risks associated with OOD actions. However, the fixed hyperparameter in policy evaluation and density-based policy improvement method limit its overall efficiency. In this paper, we propose Proj-IQL, a projective IQL algorithm enhanced with the support constraint. In the policy evaluation phase, Proj-IQL generalizes the one-step approach to a multi-step approach through vector projection, while maintaining in-sample learning and expectile regression framework. In the policy improvement phase, Proj-IQL introduces support constraint that is more aligned with the policy evaluation approach. Furthermore, we theoretically demonstrate that Proj-IQL guarantees monotonic policy improvement and enjoys a progressively more rigorous criterion for superior actions. Empirical results demonstrate the Proj-IQL achieves state-of-the-art performance on D4RL benchmarks, especially in challenging navigation domains.
☆ Multi-View Transformers for Airway-To-Lung Ratio Inference on Cardiac CT Scans: The C4R Study
The ratio of airway tree lumen to lung size (ALR), assessed at full inspiration on high resolution full-lung computed tomography (CT), is a major risk factor for chronic obstructive pulmonary disease (COPD). There is growing interest to infer ALR from cardiac CT images, which are widely available in epidemiological cohorts, to investigate the relationship of ALR to severe COVID-19 and post-acute sequelae of SARS-CoV-2 infection (PASC). Previously, cardiac scans included approximately 2/3 of the total lung volume with 5-6x greater slice thickness than high-resolution (HR) full-lung (FL) CT. In this study, we present a novel attention-based Multi-view Swin Transformer to infer FL ALR values from segmented cardiac CT scans. For the supervised training we exploit paired full-lung and cardiac CTs acquired in the Multi-Ethnic Study of Atherosclerosis (MESA). Our network significantly outperforms a proxy direct ALR inference on segmented cardiac CT scans and achieves accuracy and reproducibility comparable with a scan-rescan reproducibility of the FL ALR ground-truth.
comment: Accepted to appear in Proceedings of International Symposium on Biomedical Imaging (ISBI), 2025
☆ A Two-Stage Pretraining-Finetuning Framework for Treatment Effect Estimation with Unmeasured Confounding KDD 25
Estimating the conditional average treatment effect (CATE) from observational data plays a crucial role in areas such as e-commerce, healthcare, and economics. Existing studies mainly rely on the strong ignorability assumption that there are no unmeasured confounders, whose presence cannot be tested from observational data and can invalidate any causal conclusion. In contrast, data collected from randomized controlled trials (RCT) do not suffer from confounding, but are usually limited by a small sample size. In this paper, we propose a two-stage pretraining-finetuning (TSPF) framework using both large-scale observational data and small-scale RCT data to estimate the CATE in the presence of unmeasured confounding. In the first stage, a foundational representation of covariates is trained to estimate counterfactual outcomes through large-scale observational data. In the second stage, we propose to train an augmented representation of the covariates, which is concatenated to the foundational representation obtained in the first stage to adjust for the unmeasured confounding. To avoid overfitting caused by the small-scale RCT data in the second stage, we further propose a partial parameter initialization approach, rather than training a separate network. The superiority of our approach is validated on two public datasets with extensive experiments. The code is available at https://github.com/zhouchuanCN/KDD25-TSPF.
comment: KDD 25 Research Track
☆ PAC Learnability of Scenario Decision-Making Algorithms: Necessary and Sufficient Conditions
We study the PAC property of scenario decision-making algorithms, that is, the ability to make a decision that has an arbitrarily low risk of violating an unknown safety constraint, provided sufficiently many realizations (called scenarios) of the safety constraint are sampled. Sufficient conditions for scenario decision-making algorithms to be PAC are available in the literature, such as finiteness of the VC dimension of its associated classifier and existence of a compression scheme. We study the question of whether these sufficient conditions are also necessary. We show with counterexamples that this is not the case in general. This contrasts with binary classification learning, for which the analogous conditions are sufficient and necessary. Popular scenario decision-making algorithms, such as scenario optimization, enjoy additional properties, such as stability and consistency. We show that even under these additional assumptions the above conclusions hold. Finally, we derive a necessary condition for scenario decision-making algorithms to be PAC, inspired by the VC dimension and the so-called no-free-lunch theorem.
☆ Improved Compression Bounds for Scenario Decision Making
Scenario decision making offers a flexible way of making decision in an uncertain environment while obtaining probabilistic guarantees on the risk of failure of the decision. The idea of this approach is to draw samples of the uncertainty and make a decision based on the samples, called "scenarios". The probabilistic guarantees take the form of a bound on the probability of sampling a set of scenarios that will lead to a decision whose risk of failure is above a given maximum tolerance. This bound can be expressed as a function of the number of sampled scenarios, the maximum tolerated risk, and some intrinsic property of the problem called the "compression size". Several such bounds have been proposed in the literature under various assumptions on the problem. We propose new bounds that improve upon the existing ones without requiring stronger assumptions on the problem.
☆ Increasing Batch Size Improves Convergence of Stochastic Gradient Descent with Momentum
Stochastic gradient descent with momentum (SGDM), which is defined by adding a momentum term to SGD, has been well studied in both theory and practice. Theoretically investigated results showed that the settings of the learning rate and momentum weight affect the convergence of SGDM. Meanwhile, practical results showed that the setting of batch size strongly depends on the performance of SGDM. In this paper, we focus on mini-batch SGDM with constant learning rate and constant momentum weight, which is frequently used to train deep neural networks in practice. The contribution of this paper is showing theoretically that using a constant batch size does not always minimize the expectation of the full gradient norm of the empirical loss in training a deep neural network, whereas using an increasing batch size definitely minimizes it, that is, increasing batch size improves convergence of mini-batch SGDM. We also provide numerical results supporting our analyses, indicating specifically that mini-batch SGDM with an increasing batch size converges to stationary points faster than with a constant batch size. Python implementations of the optimizers used in the numerical experiments are available at https://anonymous.4open.science/r/momentum-increasing-batch-size-888C/.
comment: 22 pages
☆ Incrementally Learning Multiple Diverse Data Domains via Multi-Source Dynamic Expansion Model
Continual Learning seeks to develop a model capable of incrementally assimilating new information while retaining prior knowledge. However, current research predominantly addresses a straightforward learning context, wherein all data samples originate from a singular data domain. This paper shifts focus to a more complex and realistic learning environment, characterized by data samples sourced from multiple distinct domains. We tackle this intricate learning challenge by introducing a novel methodology, termed the Multi-Source Dynamic Expansion Model (MSDEM), which leverages various pre-trained models as backbones and progressively establishes new experts based on them to adapt to emerging tasks. Additionally, we propose an innovative dynamic expandable attention mechanism designed to selectively harness knowledge from multiple backbones, thereby accelerating the new task learning. Moreover, we introduce a dynamic graph weight router that strategically reuses all previously acquired parameters and representations for new task learning, maximizing the positive knowledge transfer effect, which further improves generalization performance. We conduct a comprehensive series of experiments, and the empirical findings indicate that our proposed approach achieves state-of-the-art performance.
comment: 10 pages, 5 figures
☆ ARMOR: Shielding Unlearnable Examples against Data Augmentation
Private data, when published online, may be collected by unauthorized parties to train deep neural networks (DNNs). To protect privacy, defensive noises can be added to original samples to degrade their learnability by DNNs. Recently, unlearnable examples are proposed to minimize the training loss such that the model learns almost nothing. However, raw data are often pre-processed before being used for training, which may restore the private information of protected data. In this paper, we reveal the data privacy violation induced by data augmentation, a commonly used data pre-processing technique to improve model generalization capability, which is the first of its kind as far as we are concerned. We demonstrate that data augmentation can significantly raise the accuracy of the model trained on unlearnable examples from 21.3% to 66.1%. To address this issue, we propose a defense framework, dubbed ARMOR, to protect data privacy from potential breaches of data augmentation. To overcome the difficulty of having no access to the model training process, we design a non-local module-assisted surrogate model that better captures the effect of data augmentation. In addition, we design a surrogate augmentation selection strategy that maximizes distribution alignment between augmented and non-augmented samples, to choose the optimal augmentation strategy for each class. We also use a dynamic step size adjustment algorithm to enhance the defensive noise generation process. Extensive experiments are conducted on 4 datasets and 5 data augmentation methods to verify the performance of ARMOR. Comparisons with 6 state-of-the-art defense methods have demonstrated that ARMOR can preserve the unlearnability of protected private data under data augmentation. ARMOR reduces the test accuracy of the model trained on augmented protected samples by as much as 60% more than baselines.
☆ Digital Phenotyping for Adolescent Mental Health: A Feasibility Study Employing Machine Learning to Predict Mental Health Risk From Active and Passive Smartphone Data
Background: Adolescents are particularly vulnerable to mental disorders, with over 75% of cases manifesting before the age of 25. Research indicates that only 18 to 34% of young people experiencing high levels of depression or anxiety symptoms seek support. Digital tools leveraging smartphones offer scalable and early intervention opportunities. Objective: Using a novel machine learning framework, this study evaluated the feasibility of integrating active and passive smartphone data to predict mental disorders in non-clinical adolescents. Specifically, we investigated the utility of the Mindcraft app in predicting risks for internalising and externalising disorders, eating disorders, insomnia and suicidal ideation. Methods: Participants (N=103; mean age 16.1 years) were recruited from three London schools. Participants completed the Strengths and Difficulties Questionnaire, the Eating Disorders-15 Questionnaire, Sleep Condition Indicator Questionnaire and indicated the presence/absence of suicidal ideation. They used the Mindcraft app for 14 days, contributing active data via self-reports and passive data from smartphone sensors. A contrastive pretraining phase was applied to enhance user-specific feature stability, followed by supervised fine-tuning. The model evaluation employed leave-one-subject-out cross-validation using balanced accuracy as the primary metric. Results: The integration of active and passive data achieved superior performance compared to individual data sources, with mean balanced accuracies of 0.71 for SDQ-High risk, 0.67 for insomnia, 0.77 for suicidal ideation and 0.70 for eating disorders. The contrastive learning framework stabilised daily behavioural representations, enhancing predictive robustness. This study demonstrates the potential of integrating active and passive smartphone data with advanced machine-learning techniques for predicting mental health risks.
☆ Graph Counterfactual Explainable AI via Latent Space Traversal
Explaining the predictions of a deep neural network is a nontrivial task, yet high-quality explanations for predictions are often a prerequisite for practitioners to trust these models. Counterfactual explanations aim to explain predictions by finding the ''nearest'' in-distribution alternative input whose prediction changes in a pre-specified way. However, it remains an open question how to define this nearest alternative input, whose solution depends on both the domain (e.g. images, graphs, tabular data, etc.) and the specific application considered. For graphs, this problem is complicated i) by their discrete nature, as opposed to the continuous nature of state-of-the-art graph classifiers; and ii) by the node permutation group acting on the graphs. We propose a method to generate counterfactual explanations for any differentiable black-box graph classifier, utilizing a case-specific permutation equivariant graph variational autoencoder. We generate counterfactual explanations in a continuous fashion by traversing the latent space of the autoencoder across the classification boundary of the classifier, allowing for seamless integration of discrete graph structure and continuous graph attributes. We empirically validate the approach on three graph datasets, showing that our model is consistently high-performing and more robust than the baselines.
comment: Published at Northern Lights Deep Learning Conference 2025
☆ RouteNet-Gauss: Hardware-Enhanced Network Modeling with Machine Learning
Network simulation is pivotal in network modeling, assisting with tasks ranging from capacity planning to performance estimation. Traditional approaches such as Discrete Event Simulation (DES) face limitations in terms of computational cost and accuracy. This paper introduces RouteNet-Gauss, a novel integration of a testbed network with a Machine Learning (ML) model to address these challenges. By using the testbed as a hardware accelerator, RouteNet-Gauss generates training datasets rapidly and simulates network scenarios with high fidelity to real-world conditions. Experimental results show that RouteNet-Gauss significantly reduces prediction errors by up to 95% and achieves a 488x speedup in inference time compared to state-of-the-art DES-based methods. RouteNet-Gauss's modular architecture is dynamically constructed based on the specific characteristics of the network scenario, such as topology and routing. This enables it to understand and generalize to different network configurations beyond those seen during training, including networks up to 10x larger. Additionally, it supports Temporal Aggregated Performance Estimation (TAPE), providing configurable temporal granularity and maintaining high accuracy in flow performance metrics. This approach shows promise in improving both simulation efficiency and accuracy, offering a valuable tool for network operators.
comment: 13 pages, 11 figures
☆ Deep Learning Meets Queue-Reactive: A Framework for Realistic Limit Order Book Simulation
The Queue-Reactive model introduced by Huang et al. (2015) has become a standard tool for limit order book modeling, widely adopted by both researchers and practitioners for its simplicity and effectiveness. We present the Multidimensional Deep Queue-Reactive (MDQR) model, which extends this framework in three ways: it relaxes the assumption of queue independence, enriches the state space with market features, and models the distribution of order sizes. Through a neural network architecture, the model learns complex dependencies between different price levels and adapts to varying market conditions, while preserving the interpretable point-process foundation of the original framework. Using data from the Bund futures market, we show that MDQR captures key market properties including the square-root law of market impact, cross-queue correlations, and realistic order size patterns. The model demonstrates particular strength in reproducing both conditional and stationary distributions of order sizes, as well as various stylized facts of market microstructure. The model achieves this while maintaining the computational efficiency needed for practical applications such as strategy development through reinforcement learning or realistic backtesting.
☆ A Closer Look at the Learnability of Out-of-Distribution (OOD) Detection
Machine learning algorithms often encounter different or "out-of-distribution" (OOD) data at deployment time, and OOD detection is frequently employed to detect these examples. While it works reasonably well in practice, existing theoretical results on OOD detection are highly pessimistic. In this work, we take a closer look at this problem, and make a distinction between uniform and non-uniform learnability, following PAC learning theory. We characterize under what conditions OOD detection is uniformly and non-uniformly learnable, and we show that in several cases, non-uniform learnability turns a number of negative results into positive. In all cases where OOD detection is learnable, we provide concrete learning algorithms and a sample-complexity analysis.
☆ IDEA: Image Description Enhanced CLIP-Adapter
CLIP (Contrastive Language-Image Pre-training) has attained great success in pattern recognition and computer vision. Transferring CLIP to downstream tasks (e.g. zero- or few-shot classification) is a hot topic in multimodal learning. However, current studies primarily focus on either prompt learning for text or adapter tuning for vision, without fully exploiting the complementary information and correlations among image-text pairs. In this paper, we propose an Image Description Enhanced CLIP-Adapter (IDEA) method to adapt CLIP to few-shot image classification tasks. This method captures fine-grained features by leveraging both visual features and textual descriptions of images. IDEA is a training-free method for CLIP, and it can be comparable to or even exceeds state-of-the-art models on multiple tasks. Furthermore, we introduce Trainable-IDEA (T-IDEA), which extends IDEA by adding two lightweight learnable components (i.e., a projector and a learnable latent space), further enhancing the model's performance and achieving SOTA results on 11 datasets. As one important contribution, we employ the Llama model and design a comprehensive pipeline to generate textual descriptions for images of 11 datasets, resulting in a total of 1,637,795 image-text pairs, named "IMD-11". Our code and data are released at https://github.com/FourierAI/IDEA.
☆ Deep learning for temporal super-resolution 4D Flow MRI
4D Flow Magnetic Resonance Imaging (4D Flow MRI) is a non-invasive technique for volumetric, time-resolved blood flow quantification. However, apparent trade-offs between acquisition time, image noise, and resolution limit clinical applicability. In particular, in regions of highly transient flow, coarse temporal resolution can hinder accurate capture of physiologically relevant flow variations. To overcome these issues, post-processing techniques using deep learning have shown promising results to enhance resolution post-scan using so-called super-resolution networks. However, while super-resolution has been focusing on spatial upsampling, temporal super-resolution remains largely unexplored. The aim of this study was therefore to implement and evaluate a residual network for temporal super-resolution 4D Flow MRI. To achieve this, an existing spatial network (4DFlowNet) was re-designed for temporal upsampling, adapting input dimensions, and optimizing internal layer structures. Training and testing were performed using synthetic 4D Flow MRI data originating from patient-specific in-silico models, as well as using in-vivo datasets. Overall, excellent performance was achieved with input velocities effectively denoised and temporally upsampled, with a mean absolute error (MAE) of 1.0 cm/s in an unseen in-silico setting, outperforming deterministic alternatives (linear interpolation MAE = 2.3 cm/s, sinc interpolation MAE = 2.6 cm/s). Further, the network synthesized high-resolution temporal information from unseen low-resolution in-vivo data, with strong correlation observed at peak flow frames. As such, our results highlight the potential of utilizing data-driven neural networks for temporal super-resolution 4D Flow MRI, enabling high-frame-rate flow quantification without extending acquisition times beyond clinically acceptable limits.
comment: 12 pages, 8 figures
☆ Nesterov Acceleration for Ensemble Kalman Inversion and Variants
Ensemble Kalman inversion (EKI) is a derivative-free, particle-based optimization method for solving inverse problems. It can be shown that EKI approximates a gradient flow, which allows the application of methods for accelerating gradient descent. Here, we show that Nesterov acceleration is effective in speeding up the reduction of the EKI cost function on a variety of inverse problems. We also implement Nesterov acceleration for two EKI variants, unscented Kalman inversion and ensemble transform Kalman inversion. Our specific implementation takes the form of a particle-level nudge that is demonstrably simple to couple in a black-box fashion with any existing EKI variant algorithms, comes with no additional computational expense, and with no additional tuning hyperparameters. This work shows a pathway for future research to translate advances in gradient-based optimization into advances in gradient-free Kalman optimization.
☆ Networked Agents in the Dark: Team Value Learning under Partial Observability AAMAS 2025
We propose a novel cooperative multi-agent reinforcement learning (MARL) approach for networked agents. In contrast to previous methods that rely on complete state information or joint observations, our agents must learn how to reach shared objectives under partial observability. During training, they collect individual rewards and approximate a team value function through local communication, resulting in cooperative behavior. To describe our problem, we introduce the networked dynamic partially observable Markov game framework, where agents communicate over a switching topology communication network. Our distributed method, DNA-MARL, uses a consensus mechanism for local communication and gradient descent for local computation. DNA-MARL increases the range of the possible applications of networked agents, being well-suited for real world domains that impose privacy and where the messages may not reach their recipients. We evaluate DNA-MARL across benchmark MARL scenarios. Our results highlight the superior performance of DNA-MARL over previous methods.
comment: 18 pages, 7 figures, 5 tables. Accepted as supplemental material at Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2025), Detroit, Michigan, USA, May 19 - 23, 2025, IFAAMAS
☆ Leveraging LLM Agents for Translating Network Configurations
Configuration translation is a critical and frequent task in network operations. When a network device is damaged or outdated, administrators need to replace it to maintain service continuity. The replacement devices may originate from different vendors, necessitating configuration translation to ensure seamless network operation. However, translating configurations manually is a labor-intensive and error-prone process. In this paper, we propose an intent-based framework for translating network configuration with Large Language Model (LLM) Agents. The core of our approach is an Intent-based Retrieval Augmented Generation (IRAG) module that systematically splits a configuration file into fragments, extracts intents, and generates accurate translations. We also design a two-stage verification method to validate the syntax and semantics correctness of the translated configurations. We implement and evaluate the proposed method on real-world network configurations. Experimental results show that our method achieves 97.74% syntax correctness, outperforming state-of-the-art methods in translation accuracy.
☆ MeshMask: Physics-Based Simulations with Masked Graph Neural Networks
We introduce a novel masked pre-training technique for graph neural networks (GNNs) applied to computational fluid dynamics (CFD) problems. By randomly masking up to 40\% of input mesh nodes during pre-training, we force the model to learn robust representations of complex fluid dynamics. We pair this masking strategy with an asymmetric encoder-decoder architecture and gated multi-layer perceptrons to further enhance performance. The proposed method achieves state-of-the-art results on seven CFD datasets, including a new challenging dataset of 3D intracranial aneurysm simulations with over 250,000 nodes per mesh. Moreover, it significantly improves model performance and training efficiency across such diverse range of fluid simulation tasks. We demonstrate improvements of up to 60\% in long-term prediction accuracy compared to previous best models, while maintaining similar computational costs. Notably, our approach enables effective pre-training on multiple datasets simultaneously, significantly reducing the time and data required to achieve high performance on new tasks. Through extensive ablation studies, we provide insights into the optimal masking ratio, architectural choices, and training strategies.
☆ Resource-Constrained Federated Continual Learning: What Does Matter?
Federated Continual Learning (FCL) aims to enable sequentially privacy-preserving model training on streams of incoming data that vary in edge devices by preserving previous knowledge while adapting to new data. Current FCL literature focuses on restricted data privacy and access to previously seen data while imposing no constraints on the training overhead. This is unreasonable for FCL applications in real-world scenarios, where edge devices are primarily constrained by resources such as storage, computational budget, and label rate. We revisit this problem with a large-scale benchmark and analyze the performance of state-of-the-art FCL approaches under different resource-constrained settings. Various typical FCL techniques and six datasets in two incremental learning scenarios (Class-IL and Domain-IL) are involved in our experiments. Through extensive experiments amounting to a total of over 1,000+ GPU hours, we find that, under limited resource-constrained settings, existing FCL approaches, with no exception, fail to achieve the expected performance. Our conclusions are consistent in the sensitivity analysis. This suggests that most existing FCL methods are particularly too resource-dependent for real-world deployment. Moreover, we study the performance of typical FCL techniques with resource constraints and shed light on future research directions in FCL.
comment: arXiv admin note: text overlap with arXiv:2303.11165 by other authors
☆ GRAPPA - A Hybrid Graph Neural Network for Predicting Pure Component Vapor Pressures
Although the pure component vapor pressure is one of the most important properties for designing chemical processes, no broadly applicable, sufficiently accurate, and open-source prediction method has been available. To overcome this, we have developed GRAPPA - a hybrid graph neural network for predicting vapor pressures of pure components. GRAPPA enables the prediction of the vapor pressure curve of basically any organic molecule, requiring only the molecular structure as input. The new model consists of three parts: A graph attention network for the message passing step, a pooling function that captures long-range interactions, and a prediction head that yields the component-specific parameters of the Antoine equation, from which the vapor pressure can readily and consistently be calculated for any temperature. We have trained and evaluated GRAPPA on experimental vapor pressure data of almost 25,000 pure components. We found excellent prediction accuracy for unseen components, outperforming state-of-the-art group contribution methods and other machine learning approaches in applicability and accuracy. The trained model and its code are fully disclosed, and GRAPPA is directly applicable via the interactive website ml-prop.mv.rptu.de.
comment: 38 pages, 12 figures
☆ Transformed Low-rank Adaptation via Tensor Decomposition and Its Applications to Text-to-image Models
Parameter-Efficient Fine-Tuning (PEFT) of text-to-image models has become an increasingly popular technique with many applications. Among the various PEFT methods, Low-Rank Adaptation (LoRA) and its variants have gained significant attention due to their effectiveness, enabling users to fine-tune models with limited computational resources. However, the approximation gap between the low-rank assumption and desired fine-tuning weights prevents the simultaneous acquisition of ultra-parameter-efficiency and better performance. To reduce this gap and further improve the power of LoRA, we propose a new PEFT method that combines two classes of adaptations, namely, transform and residual adaptations. In specific, we first apply a full-rank and dense transform to the pre-trained weight. This learnable transform is expected to align the pre-trained weight as closely as possible to the desired weight, thereby reducing the rank of the residual weight. Then, the residual part can be effectively approximated by more compact and parameter-efficient structures, with a smaller approximation error. To achieve ultra-parameter-efficiency in practice, we design highly flexible and effective tensor decompositions for both the transform and residual adaptations. Additionally, popular PEFT methods such as DoRA can be summarized under this transform plus residual adaptation scheme. Experiments are conducted on fine-tuning Stable Diffusion models in subject-driven and controllable generation. The results manifest that our method can achieve better performances and parameter efficiency compared to LoRA and several baselines.
☆ $\texttt{InfoHier}$: Hierarchical Information Extraction via Encoding and Embedding
Analyzing large-scale datasets, especially involving complex and high-dimensional data like images, is particularly challenging. While self-supervised learning (SSL) has proven effective for learning representations from unlabelled data, it typically focuses on flat, non-hierarchical structures, missing the multi-level relationships present in many real-world datasets. Hierarchical clustering (HC) can uncover these relationships by organizing data into a tree-like structure, but it often relies on rigid similarity metrics that struggle to capture the complexity of diverse data types. To address these we envision $\texttt{InfoHier}$, a framework that combines SSL with HC to jointly learn robust latent representations and hierarchical structures. This approach leverages SSL to provide adaptive representations, enhancing HC's ability to capture complex patterns. Simultaneously, it integrates HC loss to refine SSL training, resulting in representations that are more attuned to the underlying information hierarchy. $\texttt{InfoHier}$ has the potential to improve the expressiveness and performance of both clustering and representation learning, offering significant benefits for data analysis, management, and information retrieval.
comment: 10 pages, 4 figures
Self-supervised Transformation Learning for Equivariant Representations NeurIPS 2024
Unsupervised representation learning has significantly advanced various machine learning tasks. In the computer vision domain, state-of-the-art approaches utilize transformations like random crop and color jitter to achieve invariant representations, embedding semantically the same inputs despite transformations. However, this can degrade performance in tasks requiring precise features, such as localization or flower classification. To address this, recent research incorporates equivariant representation learning, which captures transformation-sensitive information. However, current methods depend on transformation labels and thus struggle with interdependency and complex transformations. We propose Self-supervised Transformation Learning (STL), replacing transformation labels with transformation representations derived from image pairs. The proposed method ensures transformation representation is image-invariant and learns corresponding equivariant transformations, enhancing performance without increased batch complexity. We demonstrate the approach's effectiveness across diverse classification and detection tasks, outperforming existing methods in 7 out of 11 benchmarks and excelling in detection. By integrating complex transformations like AugMix, unusable by prior equivariant methods, this approach enhances performance across tasks, underscoring its adaptability and resilience. Additionally, its compatibility with various base models highlights its flexibility and broad applicability. The code is available at https://github.com/jaemyung-u/stl.
comment: 38th Conference on Neural Information Processing Systems (NeurIPS 2024)
☆ Disentangled Interleaving Variational Encoding
Conflicting objectives present a considerable challenge in interleaving multi-task learning, necessitating the need for meticulous design and balance to ensure effective learning of a representative latent data space across all tasks without mutual negative impact. Drawing inspiration from the concept of marginal and conditional probability distributions in probability theory, we design a principled and well-founded approach to disentangle the original input into marginal and conditional probability distributions in the latent space of a variational autoencoder. Our proposed model, Deep Disentangled Interleaving Variational Encoding (DeepDIVE) learns disentangled features from the original input to form clusters in the embedding space and unifies these features via the cross-attention mechanism in the fusion stage. We theoretically prove that combining the objectives for reconstruction and forecasting fully captures the lower bound and mathematically derive a loss function for disentanglement using Na\"ive Bayes. Under the assumption that the prior is a mixture of log-concave distributions, we also establish that the Kullback-Leibler divergence between the prior and the posterior is upper bounded by a function minimized by the minimizer of the cross entropy loss, informing our adoption of radial basis functions (RBF) and cross entropy with interleaving training for DeepDIVE to provide a justified basis for convergence. Experiments on two public datasets show that DeepDIVE disentangles the original input and yields forecast accuracies better than the original VAE and comparable to existing state-of-the-art baselines.
☆ Diagonal Over-parameterization in Reproducing Kernel Hilbert Spaces as an Adaptive Feature Model: Generalization and Adaptivity
This paper introduces a diagonal adaptive kernel model that dynamically learns kernel eigenvalues and output coefficients simultaneously during training. Unlike fixed-kernel methods tied to the neural tangent kernel theory, the diagonal adaptive kernel model adapts to the structure of the truth function, significantly improving generalization over fixed-kernel methods, especially when the initial kernel is misaligned with the target. Moreover, we show that the adaptivity comes from learning the right eigenvalues during training, showing a feature learning behavior. By extending to deeper parameterization, we further show how extra depth enhances adaptability and generalization. This study combines the insights from feature learning and implicit regularization and provides new perspective into the adaptivity and generalization potential of neural networks beyond the kernel regime.
comment: arXiv admin note: text overlap with arXiv:2409.00894
☆ Investigating Parameter-Efficiency of Hybrid QuGANs Based on Geometric Properties of Generated Sea Route Graphs
The demand for artificially generated data for the development, training and testing of new algorithms is omnipresent. Quantum computing (QC), does offer the hope that its inherent probabilistic functionality can be utilised in this field of generative artificial intelligence. In this study, we use quantum-classical hybrid generative adversarial networks (QuGANs) to artificially generate graphs of shipping routes. We create a training dataset based on real shipping data and investigate to what extent QuGANs are able to learn and reproduce inherent distributions and geometric features of this data. We compare hybrid QuGANs with classical Generative Adversarial Networks (GANs), with a special focus on their parameter efficiency. Our results indicate that QuGANs are indeed able to quickly learn and represent underlying geometric properties and distributions, although they seem to have difficulties in introducing variance into the sampled data. Compared to classical GANs of greater size, measured in the number of parameters used, some QuGANs show similar result quality. Our reference to concrete use cases, such as the generation of shipping data, provides an illustrative example and demonstrate the potential and diversity in which QC can be used.
☆ SPEQ: Stabilization Phases for Efficient Q-Learning in High Update-To-Data Ratio Reinforcement Learning
A key challenge in Deep Reinforcement Learning is sample efficiency, especially in real-world applications where collecting environment interactions is expensive or risky. Recent off-policy algorithms improve sample efficiency by increasing the Update-To-Data (UTD) ratio and performing more gradient updates per environment interaction. While this improves sample efficiency, it significantly increases computational cost due to the higher number of gradient updates required. In this paper we propose a sample-efficient method to improve computational efficiency by separating training into distinct learning phases in order to exploit gradient updates more effectively. Our approach builds on top of the Dropout Q-Functions (DroQ) algorithm and alternates between an online, low UTD ratio training phase, and an offline stabilization phase. During the stabilization phase, we fine-tune the Q-functions without collecting new environment interactions. This process improves the effectiveness of the replay buffer and reduces computational overhead. Our experimental results on continuous control problems show that our method achieves results comparable to state-of-the-art, high UTD ratio algorithms while requiring 56\% fewer gradient updates and 50\% less training time than DroQ. Our approach offers an effective and computationally economical solution while maintaining the same sample efficiency as the more costly, high UTD ratio state-of-the-art.
☆ Product of Gaussian Mixture Diffusion Model for non-linear MRI Inversion
Diffusion models have recently shown remarkable results in magnetic resonance imaging reconstruction. However, the employed networks typically are black-box estimators of the (smoothed) prior score with tens of millions of parameters, restricting interpretability and increasing reconstruction time. Furthermore, parallel imaging reconstruction algorithms either rely on off-line coil sensitivity estimation, which is prone to misalignment and restricting sampling trajectories, or perform per-coil reconstruction, making the computational cost proportional to the number of coils. To overcome this, we jointly reconstruct the image and the coil sensitivities using the lightweight, parameter-efficient, and interpretable product of Gaussian mixture diffusion model as an image prior and a classical smoothness priors on the coil sensitivities. The proposed method delivers promising results while allowing for fast inference and demonstrating robustness to contrast out-of-distribution data and sampling trajectories, comparable to classical variational penalties such as total variation. Finally, the probabilistic formulation allows the calculation of the posterior expectation and pixel-wise variance.
☆ Fine-grained Spatio-temporal Event Prediction with Self-adaptive Anchor Graph SDM'25
Event prediction tasks often handle spatio-temporal data distributed in a large spatial area. Different regions in the area exhibit different characteristics while having latent correlations. This spatial heterogeneity and correlations greatly affect the spatio-temporal distributions of event occurrences, which has not been addressed by state-of-the-art models. Learning spatial dependencies of events in a continuous space is challenging due to its fine granularity and a lack of prior knowledge. In this work, we propose a novel Graph Spatio-Temporal Point Process (GSTPP) model for fine-grained event prediction. It adopts an encoder-decoder architecture that jointly models the state dynamics of spatially localized regions using neural Ordinary Differential Equations (ODEs). The state evolution is built on the foundation of a novel Self-Adaptive Anchor Graph (SAAG) that captures spatial dependencies. By adaptively localizing the anchor nodes in the space and jointly constructing the correlation edges between them, the SAAG enhances the model's ability of learning complex spatial event patterns. The proposed GSTPP model greatly improves the accuracy of fine-grained event prediction. Extensive experimental results show that our method greatly improves the prediction accuracy over existing spatio-temporal event prediction approaches.
comment: Accepted to SIAM International Conference on Data Mining 2025 (SDM'25)
☆ Joint Learning of Depth and Appearance for Portrait Image Animation
2D portrait animation has experienced significant advancements in recent years. Much research has utilized the prior knowledge embedded in large generative diffusion models to enhance high-quality image manipulation. However, most methods only focus on generating RGB images as output, and the co-generation of consistent visual plus 3D output remains largely under-explored. In our work, we propose to jointly learn the visual appearance and depth simultaneously in a diffusion-based portrait image generator. Our method embraces the end-to-end diffusion paradigm and introduces a new architecture suitable for learning this conditional joint distribution, consisting of a reference network and a channel-expanded diffusion backbone. Once trained, our framework can be efficiently adapted to various downstream applications, such as facial depth-to-image and image-to-depth generation, portrait relighting, and audio-driven talking head animation with consistent 3D output.
☆ Quantum Reservoir Computing and Risk Bounds
We propose a way to bound the generalisation errors of several classes of quantum reservoirs using the Rademacher complexity. We give specific, parameter-dependent bounds for two particular quantum reservoir classes. We analyse how the generalisation bounds scale with growing numbers of qubits. Applying our results to classes with polynomial readout functions, we find that the risk bounds converge in the number of training samples. The explicit dependence on the quantum reservoir and readout parameters in our bounds can be used to control the generalisation error to a certain extent. It should be noted that the bounds scale exponentially with the number of qubits $n$. The upper bounds on the Rademacher complexity can be applied to other reservoir classes that fulfill a few hypotheses on the quantum dynamics and the readout function.
☆ SWSC: Shared Weight for Similar Channel in LLM
Large language models (LLMs) have spurred development in multiple industries. However, the growing number of their parameters brings substantial storage and computing burdens, making it essential to explore model compression techniques for parameter reduction and easier deployment. We propose SWSC, an LLM compression method based on the concept of Shared Weight for Similar Channel. It uses the K-Means clustering algorithm to cluster model weights channel-by-channel, generating clusters with highly similar vectors within each. A representative vector from each cluster is selected to approximately replace all vectors in the cluster, significantly reducing the number of model weight parameters. However, approximate restoration will inevitably cause damage to the performance of the model. To tackle this issue, we perform singular value decomposition on the weight error values before and after compression and retain the larger singular values and their corresponding singular vectors to compensate for the accuracy. The experimental results show that our method can effectively ensure the performance of the compressed LLM even under low-precision conditions.
comment: 5pages, 3 figures, work in progress
Transformer-based Multivariate Time Series Anomaly Localization
With the growing complexity of Cyber-Physical Systems (CPS) and the integration of Internet of Things (IoT), the use of sensors for online monitoring generates large volume of multivariate time series (MTS) data. Consequently, the need for robust anomaly diagnosis in MTS is paramount to maintaining system reliability and safety. While significant advancements have been made in anomaly detection, localization remains a largely underexplored area, though crucial for intelligent decision-making. This paper introduces a novel transformer-based model for unsupervised anomaly diagnosis in MTS, with a focus on improving localization performance, through an in-depth analysis of the self-attention mechanism's learning behavior under both normal and anomalous conditions. We formulate the anomaly localization problem as a three-stage process: time-step, window, and segment-based. This leads to the development of the Space-Time Anomaly Score (STAS), a new metric inspired by the connection between transformer latent representations and space-time statistical models. STAS is designed to capture individual anomaly behaviors and inter-series dependencies, delivering enhanced localization performance. Additionally, the Statistical Feature Anomaly Score (SFAS) complements STAS by analyzing statistical features around anomalies, with their combination helping to reduce false alarms. Experiments on real world and synthetic datasets illustrate the model's superiority over state-of-the-art methods in both detection and localization tasks.
☆ A Learning Algorithm That Attains the Human Optimum in a Repeated Human-Machine Interaction Game
When humans interact with learning-based control systems, a common goal is to minimize a cost function known only to the human. For instance, an exoskeleton may adapt its assistance in an effort to minimize the human's metabolic cost-of-transport. Conventional approaches to synthesizing the learning algorithm solve an inverse problem to infer the human's cost. However, these problems can be ill-posed, hard to solve, or sensitive to problem data. Here we show a game-theoretic learning algorithm that works solely by observing human actions to find the cost minimum, avoiding the need to solve an inverse problem. We evaluate the performance of our algorithm in an extensive set of human subjects experiments, demonstrating consistent convergence to the minimum of a prescribed human cost function in scalar and multidimensional instantiations of the game. We conclude by outlining future directions for theoretical and empirical extensions of our results.
☆ CT-PatchTST: Channel-Time Patch Time-Series Transformer for Long-Term Renewable Energy Forecasting
Accurately predicting renewable energy output is crucial for the efficient integration of solar and wind power into modern energy systems. This study develops and evaluates an advanced deep learning model, Channel-Time Patch Time-Series Transformer (CT-PatchTST), to forecast the power output of photovoltaic and wind energy systems using annual offshore wind power, onshore wind power, and solar power generation data from Denmark. While the original Patch Time-Series Transformer(PatchTST) model employs a channel-independent (CI) approach, it tends to overlook inter-channel relationships during training, potentially leading to a loss of critical information. To address this limitation and further leverage the benefits of increased data granularity brought by CI, we propose CT-PatchTST. This enhanced model improves the processing of inter-channel information while maintaining the advantages of the channel-independent approach. The predictive performance of CT-PatchTST is rigorously analyzed, demonstrating its ability to provide precise and reliable energy forecasts. This work contributes to improving the predictability of renewable energy systems, supporting their broader adoption and integration into energy grids.
☆ RLHS: Mitigating Misalignment in RLHF with Hindsight Simulation
Generative AI systems like foundation models (FMs) must align well with human values to ensure their behavior is helpful and trustworthy. While Reinforcement Learning from Human Feedback (RLHF) has shown promise for optimizing model performance using human judgments, existing RLHF pipelines predominantly rely on immediate feedback, which can fail to accurately reflect the downstream impact of an interaction on users' utility. We demonstrate that feedback based on evaluators' foresight estimates of downstream consequences systematically induces Goodhart's Law dynamics, incentivizing misaligned behaviors like sycophancy and deception and ultimately degrading user outcomes. To alleviate this, we propose decoupling evaluation from prediction by refocusing RLHF on hindsight feedback. Our theoretical analysis reveals that conditioning evaluator feedback on downstream observations mitigates misalignment and improves expected human utility, even when these observations are simulated by the AI system itself. To leverage this insight in a practical alignment algorithm, we introduce Reinforcement Learning from Hindsight Simulation (RLHS), which first simulates plausible consequences and then elicits feedback to assess what behaviors were genuinely beneficial in hindsight. We apply RLHS to two widely-employed online and offline preference optimization methods -- Proximal Policy Optimization (PPO) and Direct Preference Optimization (DPO) -- and show empirically that misalignment is significantly reduced with both methods. Through an online human user study, we show that RLHS consistently outperforms RLHF in helping users achieve their goals and earns higher satisfaction ratings, despite being trained solely with simulated hindsight feedback. These results underscore the importance of focusing on long-term consequences, even simulated ones, to mitigate misalignment in RLHF.
☆ Towards Aligned Data Forgetting via Twin Machine Unlearning
Modern privacy regulations have spurred the evolution of machine unlearning, a technique enabling a trained model to efficiently forget specific training data. In prior unlearning methods, the concept of "data forgetting" is often interpreted and implemented as achieving zero classification accuracy on such data. Nevertheless, the authentic aim of machine unlearning is to achieve alignment between the unlearned model and the gold model, i.e., encouraging them to have identical classification accuracy. On the other hand, the gold model often exhibits non-zero classification accuracy due to its generalization ability. To achieve aligned data forgetting, we propose a Twin Machine Unlearning (TMU) approach, where a twin unlearning problem is defined corresponding to the original unlearning problem. Consequently, the generalization-label predictor trained on the twin problem can be transferred to the original problem, facilitating aligned data forgetting. Comprehensive empirical experiments illustrate that our approach significantly enhances the alignment between the unlearned model and the gold model.
comment: arXiv admin note: substantial text overlap with arXiv:2408.11433
☆ Neural Risk-sensitive Satisficing in Contextual Bandits
The contextual bandit problem, which is a type of reinforcement learning tasks, provides an effective framework for solving challenges in recommendation systems, such as satisfying real-time requirements, enabling personalization, addressing cold-start problems. However, contextual bandit algorithms face challenges since they need to handle large state-action spaces sequentially. These challenges include the high costs for learning and balancing exploration and exploitation, as well as large variations in performance that depend on the domain of application. To address these challenges, Tsuboya et~al. proposed the Regional Linear Risk-sensitive Satisficing (RegLinRS) algorithm. RegLinRS switches between exploration and exploitation based on how well the agent has achieved the target. However, the reward expectations in RegLinRS are linearly approximated based on features, which limits its applicability when the relationship between features and reward expectations is non-linear. To handle more complex environments, we proposed Neural Risk-sensitive Satisficing (NeuralRS), which incorporates neural networks into RegLinRS, and demonstrated its utility.
comment: Accepted by AROB-ISBC 2025
☆ OpenMLDB: A Real-Time Relational Data Feature Computation System for Online ML
Efficient and consistent feature computation is crucial for a wide range of online ML applications. Typically, feature computation is divided into two distinct phases, i.e., offline stage for model training and online stage for model serving. These phases often rely on execution engines with different interface languages and function implementations, causing significant inconsistencies. Moreover, many online ML features involve complex time-series computations (e.g., functions over varied-length table windows) that differ from standard streaming and analytical queries. Existing data processing systems (e.g., Spark, Flink, DuckDB) often incur multi-second latencies for these computations, making them unsuitable for real-time online ML applications that demand timely feature updates. This paper presents OpenMLDB, a feature computation system deployed in 4Paradigm's SageOne platform and over 100 real scenarios. Technically, OpenMLDB first employs a unified query plan generator for consistent computation results across the offline and online stages, significantly reducing feature deployment overhead. Second, OpenMLDB provides an online execution engine that resolves performance bottlenecks caused by long window computations (via pre-aggregation) and multi-table window unions (via data self-adjusting). It also provides a high-performance offline execution engine with window parallel optimization and time-aware data skew resolving. Third, OpenMLDB features a compact data format and stream-focused indexing to maximize memory usage and accelerate data access. Evaluations in testing and real workloads reveal significant performance improvements and resource savings compared to the baseline systems. The open community of OpenMLDB now has over 150 contributors and gained 1.6k stars on GitHub.
☆ Molecular Graph Contrastive Learning with Line Graph
Trapped by the label scarcity in molecular property prediction and drug design, graph contrastive learning (GCL) came forward. Leading contrastive learning works show two kinds of view generators, that is, random or learnable data corruption and domain knowledge incorporation. While effective, the two ways also lead to molecular semantics altering and limited generalization capability, respectively. To this end, we relate the \textbf{L}in\textbf{E} graph with \textbf{MO}lecular graph co\textbf{N}trastive learning and propose a novel method termed \textit{LEMON}. Specifically, by contrasting the given graph with the corresponding line graph, the graph encoder can freely encode the molecular semantics without omission. Furthermore, we present a new patch with edge attribute fusion and two local contrastive losses enhance information transmission and tackle hard negative samples. Compared with state-of-the-art (SOTA) methods for view generation, superior performance on molecular property prediction suggests the effectiveness of our proposed framework.
☆ Normalize Then Propagate: Efficient Homophilous Regularization for Few-shot Semi-Supervised Node Classification AAAI 2025
Graph Neural Networks (GNNs) have demonstrated remarkable ability in semi-supervised node classification. However, most existing GNNs rely heavily on a large amount of labeled data for training, which is labor-intensive and requires extensive domain knowledge. In this paper, we first analyze the restrictions of GNNs generalization from the perspective of supervision signals in the context of few-shot semi-supervised node classification. To address these challenges, we propose a novel algorithm named NormProp, which utilizes the homophily assumption of unlabeled nodes to generate additional supervision signals, thereby enhancing the generalization against label scarcity. The key idea is to efficiently capture both the class information and the consistency of aggregation during message passing, via decoupling the direction and Euclidean norm of node representations. Moreover, we conduct a theoretical analysis to determine the upper bound of Euclidean norm, and then propose homophilous regularization to constraint the consistency of unlabeled nodes. Extensive experiments demonstrate that NormProp achieve state-of-the-art performance under low-label rate scenarios with low computational complexity.
comment: Accepted by AAAI 2025
☆ DNMDR: Dynamic Networks and Multi-view Drug Representations for Safe Medication Recommendation
Medication Recommendation (MR) is a promising research topic which booms diverse applications in the healthcare and clinical domains. However, existing methods mainly rely on sequential modeling and static graphs for representation learning, which ignore the dynamic correlations in diverse medical events of a patient's temporal visits, leading to insufficient global structural exploration on nodes. Additionally, mitigating drug-drug interactions (DDIs) is another issue determining the utility of the MR systems. To address the challenges mentioned above, this paper proposes a novel MR method with the integration of dynamic networks and multi-view drug representations (DNMDR). Specifically, weighted snapshot sequences for dynamic heterogeneous networks are constructed based on discrete visits in temporal EHRs, and all the dynamic networks are jointly trained to gain both structural correlations in diverse medical events and temporal dependency in historical health conditions, for achieving comprehensive patient representations with both semantic features and structural relationships. Moreover, combining the drug co-occurrences and adverse drug-drug interactions (DDIs) in internal view of drug molecule structure and interactive view of drug pairs, the safe drug representations are available to obtain high-quality medication combination recommendation. Finally, extensive experiments on real world datasets are conducted for performance evaluation, and the experimental results demonstrate that the proposed DNMDR method outperforms the state-of-the-art baseline models with a large margin on various metrics such as PRAUC, Jaccard, DDI rates and so on.
☆ Adaptive Sampled Softmax with Inverted Multi-Index: Methods, Theory and Applications
The softmax function is a cornerstone of multi-class classification, integral to a wide range of machine learning applications, from large-scale retrieval and ranking models to advanced large language models. However, its computational cost grows linearly with the number of classes, which becomes prohibitively expensive in scenarios with millions or even billions of classes. The sampled softmax, which relies on self-normalized importance sampling, has emerged as a powerful alternative, significantly reducing computational complexity. Yet, its estimator remains unbiased only when the sampling distribution matches the true softmax distribution. To improve both approximation accuracy and sampling efficiency, we propose the MIDX Sampler, a novel adaptive sampling strategy based on an inverted multi-index approach. Concretely, we decompose the softmax probability into several multinomial probabilities, each associated with a specific set of codewords and the last associated with the residual score of queries, thus reducing time complexity to the number of codewords instead of the number of classes. To further boost efficiency, we replace the query-specific residual probability with a simple uniform distribution, simplifying the computation while retaining high performance. Our method is backed by rigorous theoretical analysis, addressing key concerns such as sampling bias, gradient bias, convergence rates, and generalization error bounds. The results demonstrate that a smaller divergence from the ideal softmax distribution leads to faster convergence and improved generalization. Extensive experiments on large-scale language models, sequential recommenders, and extreme multi-class classification tasks confirm that the MIDX-Sampler delivers superior effectiveness and efficiency compared to existing approaches.
comment: 40 pages
☆ MIAFEx: An Attention-based Feature Extraction Method for Medical Image Classification
Feature extraction techniques are crucial in medical image classification; however, classical feature extractors in addition to traditional machine learning classifiers often exhibit significant limitations in providing sufficient discriminative information for complex image sets. While Convolutional Neural Networks (CNNs) and Vision Transformer (ViT) have shown promise in feature extraction, they are prone to overfitting due to the inherent characteristics of medical imaging data, including small sample sizes or high intra-class variance. In this work, the Medical Image Attention-based Feature Extractor (MIAFEx) is proposed, a novel method that employs a learnable refinement mechanism to enhance the classification token within the Transformer encoder architecture. This mechanism adjusts the token based on learned weights, improving the extraction of salient features and enhancing the model's adaptability to the challenges presented by medical imaging data. The MIAFEx output features quality is compared against classical feature extractors using traditional and hybrid classifiers. Also, the performance of these features is compared against modern CNN and ViT models in classification tasks, demonstrating its superiority in accuracy and robustness across multiple complex classification medical imaging datasets. This advantage is particularly pronounced in scenarios with limited training data, where traditional and modern models often struggle to generalize effectively. The source code of this proposal can be found at https://github.com/Oscar-RamosS/Medical-Image-Attention-based-Feature-Extractor-MIAFEx
comment: In preparation for Journal Submission
☆ ANSR-DT: An Adaptive Neuro-Symbolic Learning and Reasoning Framework for Digital Twins
In this paper, we propose an Adaptive Neuro-Symbolic Learning Framework for digital twin technology called ``ANSR-DT." Our approach combines pattern recognition algorithms with reinforcement learning and symbolic reasoning to enable real-time learning and adaptive intelligence. This integration enhances the understanding of the environment and promotes continuous learning, leading to better and more effective decision-making in real-time for applications that require human-machine collaboration. We evaluated the \textit{ANSR-DT} framework for its ability to learn and adapt to dynamic patterns, observing significant improvements in decision accuracy, reliability, and interpretability when compared to existing state-of-the-art methods. However, challenges still exist in extracting and integrating symbolic rules in complex environments, which limits the full potential of our framework in heterogeneous settings. Moreover, our ongoing research aims to address this issue in the future by ensuring seamless integration of neural models at large. In addition, our open-source implementation promotes reproducibility and encourages future research to build on our foundational work.
☆ LAMS: LLM-Driven Automatic Mode Switching for Assistive Teleoperation
Teleoperating high degrees-of-freedom (DoF) robotic manipulators via low-DoF controllers like joysticks often requires frequent switching between control modes, where each mode maps controller movements to specific robot actions. Manually performing this frequent switching can make teleoperation cumbersome and inefficient. On the other hand, existing automatic mode-switching solutions, such as heuristic-based or learning-based methods, are often task-specific and lack generalizability. In this paper, we introduce LLM-Driven Automatic Mode Switching (LAMS), a novel approach that leverages Large Language Models (LLMs) to automatically switch control modes based on task context. Unlike existing methods, LAMS requires no prior task demonstrations and incrementally improves by integrating user-generated mode-switching examples. We validate LAMS through an ablation study and a user study with 10 participants on complex, long-horizon tasks, demonstrating that LAMS effectively reduces manual mode switches, is preferred over alternative methods, and improves performance over time. The project website with supplementary materials is at https://lams-assistance.github.io/.
☆ Reinforcement Learning-Enhanced Procedural Generation for Dynamic Narrative-Driven AR Experiences
Procedural Content Generation (PCG) is widely used to create scalable and diverse environments in games. However, existing methods, such as the Wave Function Collapse (WFC) algorithm, are often limited to static scenarios and lack the adaptability required for dynamic, narrative-driven applications, particularly in augmented reality (AR) games. This paper presents a reinforcement learning-enhanced WFC framework designed for mobile AR environments. By integrating environment-specific rules and dynamic tile weight adjustments informed by reinforcement learning (RL), the proposed method generates maps that are both contextually coherent and responsive to gameplay needs. Comparative evaluations and user studies demonstrate that the framework achieves superior map quality and delivers immersive experiences, making it well-suited for narrative-driven AR games. Additionally, the method holds promise for broader applications in education, simulation training, and immersive extended reality (XR) experiences, where dynamic and adaptive environments are critical.
comment: Number of pages: 13, Number of figures: 4. Accepted for presentation at GRAPP 2025 - 20th International Conference on Computer Graphics Theory and Applications (for additional details on the conference visit https://grapp.scitevents.org). Disclaimer: This preprint may differ from the final version published in the conference proceedings
☆ A Theory of Optimistically Universal Online Learnability for General Concept Classes NeurIPS 2024
We provide a full characterization of the concept classes that are optimistically universally online learnable with $\{0, 1\}$ labels. The notion of optimistically universal online learning was defined in [Hanneke, 2021] in order to understand learnability under minimal assumptions. In this paper, following the philosophy behind that work, we investigate two questions, namely, for every concept class: (1) What are the minimal assumptions on the data process admitting online learnability? (2) Is there a learning algorithm which succeeds under every data process satisfying the minimal assumptions? Such an algorithm is said to be optimistically universal for the given concept class. We resolve both of these questions for all concept classes, and moreover, as part of our solution, we design general learning algorithms for each case. Finally, we extend these algorithms and results to the agnostic case, showing an equivalence between the minimal assumptions on the data process for learnability in the agnostic and realizable cases, for every concept class, as well as the equivalence of optimistically universal learnability.
comment: NeurIPS 2024
☆ OMEGA: A Low-Latency GNN Serving System for Large Graphs
Graph Neural Networks (GNNs) have been widely adopted for their ability to compute expressive node representations in graph datasets. However, serving GNNs on large graphs is challenging due to the high communication, computation, and memory overheads of constructing and executing computation graphs, which represent information flow across large neighborhoods. Existing approximation techniques in training can mitigate the overheads but, in serving, still lead to high latency and/or accuracy loss. To this end, we propose OMEGA, a system that enables low-latency GNN serving for large graphs with minimal accuracy loss through two key ideas. First, OMEGA employs selective recomputation of precomputed embeddings, which allows for reusing precomputed computation subgraphs while selectively recomputing a small fraction to minimize accuracy loss. Second, we develop computation graph parallelism, which reduces communication overhead by parallelizing the creation and execution of computation graphs across machines. Our evaluation with large graph datasets and GNN models shows that OMEGA significantly outperforms state-of-the-art techniques.
☆ Homophily-aware Heterogeneous Graph Contrastive Learning
Heterogeneous graph pre-training (HGP) has demonstrated remarkable performance across various domains. However, the issue of heterophily in real-world heterogeneous graphs (HGs) has been largely overlooked. To bridge this research gap, we proposed a novel heterogeneous graph contrastive learning framework, termed HGMS, which leverages connection strength and multi-view self-expression to learn homophilous node representations. Specifically, we design a heterogeneous edge dropping augmentation strategy that enhances the homophily of augmented views. Moreover, we introduce a multi-view self-expressive learning method to infer the homophily between nodes. In practice, we develop two approaches to solve the self-expressive matrix. The solved self-expressive matrix serves as an additional augmented view to provide homophilous information and is used to identify false negatives in contrastive loss. Extensive experimental results demonstrate the superiority of HGMS across different downstream tasks.
☆ Complexity Control Facilitates Reasoning-Based Compositional Generalization in Transformers
Transformers have demonstrated impressive capabilities across various tasks, yet their performance on compositional problems remains a subject of debate. In this study, we investigate the internal mechanisms underlying Transformers' behavior in compositional tasks. We find that complexity control strategies significantly influence whether the model learns primitive-level rules that generalize out-of-distribution (reasoning-based solutions) or relies solely on memorized mappings (memory-based solutions). By applying masking strategies to the model's information circuits and employing multiple complexity metrics, we reveal distinct internal working mechanisms associated with different solution types. Further analysis reveals that reasoning-based solutions exhibit a lower complexity bias, which aligns with the well-studied neuron condensation phenomenon. This lower complexity bias is hypothesized to be the key factor enabling these solutions to learn reasoning rules. We validate these conclusions across multiple real-world datasets, including image generation and natural language processing tasks, confirming the broad applicability of our findings.
comment: Mistakenly submitted as a replacement to 2405.05409v4
☆ Mitigating Domain Shift in Federated Learning via Intra- and Inter-Domain Prototypes
Federated Learning (FL) has emerged as a decentralized machine learning technique, allowing clients to train a global model collaboratively without sharing private data. However, most FL studies ignore the crucial challenge of heterogeneous domains where each client has a distinct feature distribution, which is common in real-world scenarios. Prototype learning, which leverages the mean feature vectors within the same classes, has become a prominent solution for federated learning under domain skew. However, existing federated prototype learning methods only consider inter-domain prototypes on the server and overlook intra-domain characteristics. In this work, we introduce a novel federated prototype learning method, namely I$^2$PFL, which incorporates $\textbf{I}$ntra-domain and $\textbf{I}$nter-domain $\textbf{P}$rototypes, to mitigate domain shifts and learn a generalized global model across multiple domains in federated learning. To construct intra-domain prototypes, we propose feature alignment with MixUp-based augmented prototypes to capture the diversity of local domains and enhance the generalization of local features. Additionally, we introduce a reweighting mechanism for inter-domain prototypes to generate generalized prototypes to provide inter-domain knowledge and reduce domain skew across multiple clients. Extensive experiments on the Digits, Office-10, and PACS datasets illustrate the superior performance of our method compared to other baselines.
comment: 13 pages, 9 figures, 10 tables
☆ Learning Hyperplane Tree: A Piecewise Linear and Fully Interpretable Decision-making Framework
This paper introduces a novel tree-based model, Learning Hyperplane Tree (LHT), which outperforms state-of-the-art (SOTA) tree models for classification tasks on several public datasets. The structure of LHT is simple and efficient: it partitions the data using several hyperplanes to progressively distinguish between target and non-target class samples. Although the separation is not perfect at each stage, LHT effectively improves the distinction through successive partitions. During testing, a sample is classified by evaluating the hyperplanes defined in the branching blocks and traversing down the tree until it reaches the corresponding leaf block. The class of the test sample is then determined using the piecewise linear membership function defined in the leaf blocks, which is derived through least-squares fitting and fuzzy logic. LHT is highly transparent and interpretable--at each branching block, the contribution of each feature to the classification can be clearly observed.
☆ Score-based 3D molecule generation with neural fields NeurIPS 2024
We introduce a new representation for 3D molecules based on their continuous atomic density fields. Using this representation, we propose a new model based on walk-jump sampling for unconditional 3D molecule generation in the continuous space using neural fields. Our model, FuncMol, encodes molecular fields into latent codes using a conditional neural field, samples noisy codes from a Gaussian-smoothed distribution with Langevin MCMC (walk), denoises these samples in a single step (jump), and finally decodes them into molecular fields. FuncMol performs all-atom generation of 3D molecules without assumptions on the molecular structure and scales well with the size of molecules, unlike most approaches. Our method achieves competitive results on drug-like molecules and easily scales to macro-cyclic peptides, with at least one order of magnitude faster sampling. The code is available at https://github.com/prescient-design/funcmol.
comment: NeurIPS 2024
☆ Exploring the Efficacy of Meta-Learning: Unveiling Superior Data Diversity Utilization of MAML Over Pre-training
Currently, data and model size dominate the narrative in the training of super-large, powerful models. However, there has been a lack of exploration on the effect of other attributes of the training dataset on model performance. We hypothesize that dataset diversity can impact the performance of vision models. Our study shows positive correlations between test set accuracy and data diversity, providing an argument for furthering the research of dataset attributes beyond size. We analyzed pre-training and model-agnostic meta-learning methods on twelve popular visual datasets (e.g., Omniglot, CIFAR-FS, Aircraft) and five model configurations, including MAML variants with different numbers of inner gradient steps and supervised learning. We show moderate to strong positive correlations (R-squared: 0.15-0.42) between accuracy and data diversity and weaker but significant correlations (R-squared: ~0.2) between loss and diversity. These findings support our hypothesis and demonstrate a promising way for a deeper exploration of how formal data diversity influences model performance. This initial study highlights the potential of (Task2Vec) data diversity as a valuable measure in the rapidly evolving field of large-scale learning and emphasizes that understanding the dataset is key to building more powerful and generalizable models.
☆ SuperSAM: Crafting a SAM Supernetwork via Structured Pruning and Unstructured Parameter Prioritization
Neural Architecture Search (NAS) is a powerful approach of automating the design of efficient neural architectures. In contrast to traditional NAS methods, recently proposed one-shot NAS methods prove to be more efficient in performing NAS. One-shot NAS works by generating a singular weight-sharing supernetwork that acts as a search space (container) of subnetworks. Despite its achievements, designing the one-shot search space remains a major challenge. In this work we propose a search space design strategy for Vision Transformer (ViT)-based architectures. In particular, we convert the Segment Anything Model (SAM) into a weight-sharing supernetwork called SuperSAM. Our approach involves automating the search space design via layer-wise structured pruning and parameter prioritization. While the structured pruning applies probabilistic removal of certain transformer layers, parameter prioritization performs weight reordering and slicing of MLP-blocks in the remaining layers. We train supernetworks on several datasets using the sandwich rule. For deployment, we enhance subnetwork discovery by utilizing a program autotuner to identify efficient subnetworks within the search space. The resulting subnetworks are 30-70% smaller in size compared to the original pre-trained SAM ViT-B, yet outperform the pretrained model. Our work introduces a new and effective method for ViT NAS search-space design.
☆ Scalable Bayesian Physics-Informed Kolmogorov-Arnold Networks
Uncertainty quantification (UQ) plays a pivotal role in scientific machine learning, especially when surrogate models are used to approximate complex systems. Although multilayer perceptions (MLPs) are commonly employed as surrogates, they often suffer from overfitting due to their large number of parameters. Kolmogorov-Arnold networks (KANs) offer an alternative solution with fewer parameters. However, gradient-based inference methods, such as Hamiltonian Monte Carlo (HMC), may result in computational inefficiency when applied to KANs, especially for large-scale datasets, due to the high cost of back-propagation.To address these challenges, we propose a novel approach, combining the dropout Tikhonov ensemble Kalman inversion (DTEKI) with Chebyshev KANs. This gradient-free method effectively mitigates overfitting and enhances numerical stability. Additionally, we incorporate the active subspace method to reduce the parameter-space dimensionality, allowing us to improve the accuracy of predictions and obtain more reliable uncertainty estimates.Extensive experiments demonstrate the efficacy of our approach in various test cases, including scenarios with large datasets and high noise levels. Our results show that the new method achieves comparable or better accuracy, much higher efficiency as well as stability compared to HMC, in addition to scalability. Moreover, by leveraging the low-dimensional parameter subspace, our method preserves prediction accuracy while substantially reducing further the computational cost.
☆ Testing Noise Assumptions of Learning Algorithms
We pose a fundamental question in computational learning theory: can we efficiently test whether a training set satisfies the assumptions of a given noise model? This question has remained unaddressed despite decades of research on learning in the presence of noise. In this work, we show that this task is tractable and present the first efficient algorithm to test various noise assumptions on the training data. To model this question, we extend the recently proposed testable learning framework of Rubinfeld and Vasilyan (2023) and require a learner to run an associated test that satisfies the following two conditions: (1) whenever the test accepts, the learner outputs a classifier along with a certificate of optimality, and (2) the test must pass for any dataset drawn according to a specified modeling assumption on both the marginal distribution and the noise model. We then consider the problem of learning halfspaces over Gaussian marginals with Massart noise (where each label can be flipped with probability less than $1/2$ depending on the input features), and give a fully-polynomial time testable learning algorithm. We also show a separation between the classical setting of learning in the presence of structured noise and testable learning. In fact, for the simple case of random classification noise (where each label is flipped with fixed probability $\eta = 1/2$), we show that testable learning requires super-polynomial time while classical learning is trivial.
☆ Patch-aware Vector Quantized Codebook Learning for Unsupervised Visual Defect Detection ICTAI 2024
Unsupervised visual defect detection is critical in industrial applications, requiring a representation space that captures normal data features while detecting deviations. Achieving a balance between expressiveness and compactness is challenging; an overly expressive space risks inefficiency and mode collapse, impairing detection accuracy. We propose a novel approach using an enhanced VQ-VAE framework optimized for unsupervised defect detection. Our model introduces a patch-aware dynamic code assignment scheme, enabling context-sensitive code allocation to optimize spatial representation. This strategy enhances normal-defect distinction and improves detection accuracy during inference. Experiments on MVTecAD, BTAD, and MTSD datasets show our method achieves state-of-the-art performance.
comment: 7 pages, Accepted to 36th IEEE ICTAI 2024
☆ Enhancing Graph Representation Learning with Localized Topological Features
Representation learning on graphs is a fundamental problem that can be crucial in various tasks. Graph neural networks, the dominant approach for graph representation learning, are limited in their representation power. Therefore, it can be beneficial to explicitly extract and incorporate high-order topological and geometric information into these models. In this paper, we propose a principled approach to extract the rich connectivity information of graphs based on the theory of persistent homology. Our method utilizes the topological features to enhance the representation learning of graph neural networks and achieve state-of-the-art performance on various node classification and link prediction benchmarks. We also explore the option of end-to-end learning of the topological features, i.e., treating topological computation as a differentiable operator during learning. Our theoretical analysis and empirical study provide insights and potential guidelines for employing topological features in graph learning tasks.
comment: Accepted in JMLR 2025
☆ Generative AI Takes a Statistics Exam: A Comparison of Performance between ChatGPT3.5, ChatGPT4, and ChatGPT4o-mini
Many believe that use of generative AI as a private tutor has the potential to shrink access and achievement gaps between students and schools with abundant resources versus those with fewer resources. Shrinking the gap is possible only if paid and free versions of the platforms perform with the same accuracy. In this experiment, we investigate the performance of GPT versions 3.5, 4.0, and 4o-mini on the same 16-question statistics exam given to a class of first-year graduate students. While we do not advocate using any generative AI platform to complete an exam, the use of exam questions allows us to explore aspects of ChatGPT's responses to typical questions that students might encounter in a statistics course. Results on accuracy indicate that GPT 3.5 would fail the exam, GPT4 would perform well, and GPT4o-mini would perform somewhere in between. While we acknowledge the existence of other Generative AI/LLMs, our discussion concerns only ChatGPT because it is the most widely used platform on college campuses at this time. We further investigate differences among the AI platforms in the answers for each problem using methods developed for text analytics, such as reading level evaluation and topic modeling. Results indicate that GPT3.5 and 4o-mini have characteristics that are more similar than either of them have with GPT4.
comment: 24 pages, 2 figures, 3 tables. Submitted for publication August, 2024; revision submitted January 2025
☆ Attention is All You Need Until You Need Retention
This work introduces a novel Retention Layer mechanism for Transformer based architectures, addressing their inherent lack of intrinsic retention capabilities. Unlike human cognition, which can encode and dynamically recall symbolic templates, Generative Pretrained Transformers rely solely on fixed pretrained weights and ephemeral context windows, limiting their adaptability. The proposed Retention Layer incorporates a persistent memory module capable of real time data population, dynamic recall, and guided output generation. This enhancement allows models to store, update, and reuse observed patterns across sessions, enabling incremental learning and bridging the gap between static pretraining and dynamic, context sensitive adaptation. The Retention Layer design parallels social learning processes, encompassing attention, retention, reproduction, and motivation stages. Technically, it integrates a memory attention mechanism and episodic buffers to manage memory scalability, mitigate overfitting, and ensure efficient recall. Applications span adaptive personal assistants, real time fraud detection, autonomous robotics, content moderation, and healthcare diagnostics. In each domain, the retention mechanism enables systems to learn incrementally, personalize outputs, and respond to evolving real world challenges effectively. By emulating key aspects of human learning, this retention enhanced architecture fosters a more fluid and responsive AI paradigm, paving the way for dynamic, session aware models that extend the capabilities of traditional Transformers into domains requiring continual adaptation.
☆ Towards Understanding Extrapolation: a Causal Lens NeurIPS 2024
Canonical work handling distribution shifts typically necessitates an entire target distribution that lands inside the training distribution. However, practical scenarios often involve only a handful of target samples, potentially lying outside the training support, which requires the capability of extrapolation. In this work, we aim to provide a theoretical understanding of when extrapolation is possible and offer principled methods to achieve it without requiring an on-support target distribution. To this end, we formulate the extrapolation problem with a latent-variable model that embodies the minimal change principle in causal mechanisms. Under this formulation, we cast the extrapolation problem into a latent-variable identification problem. We provide realistic conditions on shift properties and the estimation objectives that lead to identification even when only one off-support target sample is available, tackling the most challenging scenarios. Our theory reveals the intricate interplay between the underlying manifold's smoothness and the shift properties. We showcase how our theoretical results inform the design of practical adaptation algorithms. Through experiments on both synthetic and real-world data, we validate our theoretical findings and their practical implications.
comment: NeurIPS 2024
☆ AutoLoop: Fast Visual SLAM Fine-tuning through Agentic Curriculum Learning
Current visual SLAM systems face significant challenges in balancing computational efficiency with robust loop closure handling. Traditional approaches require careful manual tuning and incur substantial computational overhead, while learning-based methods either lack explicit loop closure capabilities or implement them through computationally expensive methods. We present AutoLoop, a novel approach that combines automated curriculum learning with efficient fine-tuning for visual SLAM systems. Our method employs a DDPG (Deep Deterministic Policy Gradient) agent to dynamically adjust loop closure weights during training, eliminating the need for manual hyperparameter search while significantly reducing the required training steps. The approach pre-computes potential loop closure pairs offline and leverages them through an agent-guided curriculum, allowing the model to adapt efficiently to new scenarios. Experiments conducted on TartanAir for training and validated across multiple benchmarks including KITTI, EuRoC, ICL-NUIM and TUM RGB-D demonstrate that AutoLoop achieves comparable or superior performance while reducing training time by an order of magnitude compared to traditional approaches. AutoLoop provides a practical solution for rapid adaptation of visual SLAM systems, automating the weight tuning process that traditionally requires multiple manual iterations. Our results show that this automated curriculum strategy not only accelerates training but also maintains or improves the model's performance across diverse environmental conditions.
☆ Towards Federated Multi-Armed Bandit Learning for Content Dissemination using Swarm of UAVs
This paper introduces an Unmanned Aerial Vehicle - enabled content management architecture that is suitable for critical content access in communities of users that are communication-isolated during diverse types of disaster scenarios. The proposed architecture leverages a hybrid network of stationary anchor UAVs and mobile Micro-UAVs for ubiquitous content dissemination. The anchor UAVs are equipped with both vertical and lateral communication links, and they serve local users, while the mobile micro-ferrying UAVs extend coverage across communities with increased mobility. The focus is on developing a content dissemination system that dynamically learns optimal caching policies to maximize content availability. The core innovation is an adaptive content dissemination framework based on distributed Federated Multi-Armed Bandit learning. The goal is to optimize UAV content caching decisions based on geo-temporal content popularity and user demand variations. A Selective Caching Algorithm is also introduced to reduce redundant content replication by incorporating inter-UAV information sharing. This method strategically preserves the uniqueness in user preferences while amalgamating the intelligence across a distributed learning system. This approach improves the learning algorithm's ability to adapt to diverse user preferences. Functional verification and performance evaluation confirm the proposed architecture's utility across different network sizes, UAV swarms, and content popularity patterns.
comment: 25 pages, 11 figures, 1 table, 4 algorithms, journal
☆ Gradient Descent Converges Linearly to Flatter Minima than Gradient Flow in Shallow Linear Networks
We study the gradient descent (GD) dynamics of a depth-2 linear neural network with a single input and output. We show that GD converges at an explicit linear rate to a global minimum of the training loss, even with a large stepsize -- about $2/\textrm{sharpness}$. It still converges for even larger stepsizes, but may do so very slowly. We also characterize the solution to which GD converges, which has lower norm and sharpness than the gradient flow solution. Our analysis reveals a trade off between the speed of convergence and the magnitude of implicit regularization. This sheds light on the benefits of training at the ``Edge of Stability'', which induces additional regularization by delaying convergence and may have implications for training more complex models.
comment: 23 pages, 3 figures
♻ ☆ Delay Sensitive Hierarchical Federated Learning with Stochastic Local Updates
The impact of local averaging on the performance of federated learning (FL) systems is studied in the presence of communication delay between the clients and the parameter server. To minimize the effect of delay, clients are assigned into different groups, each having its own local parameter server (LPS) that aggregates its clients' models. The groups' models are then aggregated at a global parameter server (GPS) that only communicates with the LPSs. Such setting is known as hierarchical FL (HFL). Unlike most works in the literature, the number of local and global communication rounds in our work is randomly determined by the (different) delays experienced by each group of clients. Specifically, the number of local averaging rounds is tied to a wall-clock time period coined the sync time $S$, after which the LPSs synchronize their models by sharing them with the GPS. Such sync time $S$ is then reapplied until a global wall-clock time is exhausted. First, an upper bound on the deviation between the updated model at each LPS with respect to that available at the GPS is derived. This is then used as a tool to derive the convergence analysis of our proposed delay-sensitive HFL algorithm, first at each LPS individually, and then at the GPS. Our theoretical convergence bound showcases the effects of the whole system's parameters, including the number of groups, the number of clients per group, and the value of $S$. Our results show that the value of $S$ should be carefully chosen, especially since it implicitly governs how the delay statistics affect the performance of HFL in situations where training time is restricted.
comment: To appear in the IEEE Transactions on Cognitive Communications and Networking
♻ ☆ Reward Machines for Deep RL in Noisy and Uncertain Environments
Reward Machines provide an automaton-inspired structure for specifying instructions, safety constraints, and other temporally extended reward-worthy behaviour. By exposing the underlying structure of a reward function, they enable the decomposition of an RL task, leading to impressive gains in sample efficiency. Although Reward Machines and similar formal specifications have a rich history of application towards sequential decision-making problems, they critically rely on a ground-truth interpretation of the domain-specific vocabulary that forms the building blocks of the reward function--such ground-truth interpretations are elusive in the real world due in part to partial observability and noisy sensing. In this work, we explore the use of Reward Machines for Deep RL in noisy and uncertain environments. We characterize this problem as a POMDP and propose a suite of RL algorithms that exploit task structure under uncertain interpretation of the domain-specific vocabulary. Through theory and experiments, we expose pitfalls in naive approaches to this problem while simultaneously demonstrating how task structure can be successfully leveraged under noisy interpretations of the vocabulary.
♻ ☆ A General Framework for Inference-time Scaling and Steering of Diffusion Models
Diffusion models produce impressive results in modalities ranging from images and video to protein design and text. However, generating samples with user-specified properties remains a challenge. Recent research proposes fine-tuning models to maximize rewards that capture desired properties, but these methods require expensive training and are prone to mode collapse. In this work, we propose Feynman Kac (FK) steering, an inference-time framework for steering diffusion models with reward functions. FK steering works by sampling a system of multiple interacting diffusion processes, called particles, and resampling particles at intermediate steps based on scores computed using functions called potentials. Potentials are defined using rewards for intermediate states and are selected such that a high value indicates that the particle will yield a high-reward sample. We explore various choices of potentials, intermediate rewards, and samplers. We evaluate FK steering on text-to-image and text diffusion models. For steering text-to-image models with a human preference reward, we find that FK steering a 0.8B parameter model outperforms a 2.6B parameter fine-tuned model on prompt fidelity, with faster sampling and no training. For steering text diffusion models with rewards for text quality and specific text attributes, we find that FK steering generates lower perplexity, more linguistically acceptable outputs and enables gradient-free control of attributes like toxicity. Our results demonstrate that inference-time scaling and steering of diffusion models, even with off-the-shelf rewards, can provide significant sample quality gains and controllability benefits. Code is available at https://github.com/zacharyhorvitz/Fk-Diffusion-Steering .
♻ ☆ Optimal Federated Learning for Functional Mean Estimation under Heterogeneous Privacy Constraints
Federated learning (FL) is a distributed machine learning technique designed to preserve data privacy and security, and it has gained significant importance due to its broad range of applications. This paper addresses the problem of optimal functional mean estimation from discretely sampled data in a federated setting. We consider a heterogeneous framework where the number of individuals, measurements per individual, and privacy parameters vary across one or more servers, under both common and independent design settings. In the common design setting, the same design points are measured for each individual, whereas in the independent design, each individual has their own random collection of design points. Within this framework, we establish minimax upper and lower bounds for the estimation error of the underlying mean function, highlighting the nuanced differences between common and independent designs under distributed privacy constraints. We propose algorithms that achieve the optimal trade-off between privacy and accuracy and provide optimality results that quantify the fundamental limits of private functional mean estimation across diverse distributed settings. These results characterize the cost of privacy and offer practical insights into the potential for privacy-preserving statistical analysis in federated environments.
comment: 54 pages: 25 page article and 29 pages of appendix
♻ ☆ Debiasing Synthetic Data Generated by Deep Generative Models NeurIPS 2024
While synthetic data hold great promise for privacy protection, their statistical analysis poses significant challenges that necessitate innovative solutions. The use of deep generative models (DGMs) for synthetic data generation is known to induce considerable bias and imprecision into synthetic data analyses, compromising their inferential utility as opposed to original data analyses. This bias and uncertainty can be substantial enough to impede statistical convergence rates, even in seemingly straightforward analyses like mean calculation. The standard errors of such estimators then exhibit slower shrinkage with sample size than the typical 1 over root-$n$ rate. This complicates fundamental calculations like p-values and confidence intervals, with no straightforward remedy currently available. In response to these challenges, we propose a new strategy that targets synthetic data created by DGMs for specific data analyses. Drawing insights from debiased and targeted machine learning, our approach accounts for biases, enhances convergence rates, and facilitates the calculation of estimators with easily approximated large sample variances. We exemplify our proposal through a simulation study on toy data and two case studies on real-world data, highlighting the importance of tailoring DGMs for targeted data analysis. This debiasing strategy contributes to advancing the reliability and applicability of synthetic data in statistical inference.
comment: Accepted for the 38th Conference on Neural Information Processing Systems (NeurIPS 2024), joint first authors
♻ ☆ Customizable LLM-Powered Chatbot for Behavioral Science Research
The rapid advancement of Artificial Intelligence has resulted in the advent of Large Language Models (LLMs) with the capacity to produce text that closely resembles human communication. These models have been seamlessly integrated into diverse applications, enabling interactive and responsive communication across multiple platforms. The potential utility of chatbots transcends these traditional applications, particularly in research contexts, wherein they can offer valuable insights and facilitate the design of innovative experiments. In this study, we present a Customizable LLM-Powered Chatbot (CLPC), a web-based chatbot system designed to assist in behavioral science research. The system is meticulously designed to function as an experimental instrument rather than a conventional chatbot, necessitating users to input a username and experiment code upon access. This setup facilitates precise data cross-referencing, thereby augmenting the integrity and applicability of the data collected for research purposes. It can be easily expanded to accommodate new basic events as needed; and it allows researchers to integrate their own logging events without the necessity of implementing a separate logging mechanism. It is worth noting that our system was built to assist primarily behavioral science research but is not limited to it, it can easily be adapted to assist information retrieval research or interacting with chat bot agents in general.
♻ ☆ A Discrete-sequence Dataset for Evaluating Online Unsupervised Anomaly Detection Approaches for Multivariate Time Series
Benchmarking anomaly detection approaches for multivariate time series is challenging due to the lack of high-quality datasets. Current publicly available datasets are too small, not diverse and feature trivial anomalies, which hinders measurable progress in this research area. We propose a solution: a diverse, extensive, and non-trivial dataset generated via state-of-the-art simulation tools that reflects realistic behaviour of an automotive powertrain, including its multivariate, dynamic and variable-state properties. To cater for both unsupervised and semi-supervised anomaly detection settings, as well as time series generation and forecasting, we make different versions of the dataset available, where training and test subsets are offered in contaminated and clean versions, depending on the task. We also provide baseline results from a small selection of approaches based on deterministic and variational autoencoders, as well as a non-parametric approach. As expected, the baseline experimentation shows that the approaches trained on the semi-supervised version of the dataset outperform their unsupervised counterparts, highlighting a need for approaches more robust to contaminated training data.
comment: Submitted to the IEEE Transactions on Reliability journal
♻ ☆ Identifying Spurious Correlations using Counterfactual Alignment
Models driven by spurious correlations often yield poor generalization performance. We propose the counterfactual (CF) alignment method to detect and quantify spurious correlations of black box classifiers. Our methodology is based on counterfactual images generated with respect to one classifier being input into other classifiers to see if they also induce changes in the outputs of these classifiers. The relationship between these responses can be quantified and used to identify specific instances where a spurious correlation exists. This is validated by observing intuitive trends in face-attribute and waterbird classifiers, as well as by fabricating spurious correlations and detecting their presence, both visually and quantitatively. Furthermore, utilizing the CF alignment method, we demonstrate that we can evaluate robust optimization methods (GroupDRO, JTT, and FLAC) by detecting a reduction in spurious correlations.
comment: Accepted to Transactions on Machine Learning Research (TMLR), Code: https://github.com/ieee8023/latentshift
♻ ☆ PACE: Marrying generalization in PArameter-efficient fine-tuning with Consistency rEgularization NeurIPS 2024
Parameter-Efficient Fine-Tuning (PEFT) effectively adapts pre-trained transformers to downstream tasks. However, the optimization of tasks performance often comes at the cost of generalizability in fine-tuned models. To address this issue, we theoretically connect smaller weight gradient norms during training and larger datasets to the improvements in model generalization. Motivated by this connection, we propose reducing gradient norms for enhanced generalization and aligning fine-tuned model with the pre-trained counterpart to retain knowledge from large-scale pre-training data. Yet, naive alignment does not guarantee gradient reduction and can potentially cause gradient explosion, complicating efforts to manage gradients. To address such an issue, we propose PACE, marrying generalization of PArameter-efficient fine-tuning with Consistency rEgularization. We perturb features learned from the adapter with the multiplicative noise and ensure the fine-tuned model remains consistent for same sample under different perturbations. Theoretical analysis shows that PACE not only implicitly regularizes gradients for enhanced generalization, but also implicitly aligns the fine-tuned and pre-trained models to retain knowledge. Experimental evidence supports our theories. PACE surpasses existing PEFT methods in visual adaptation tasks (VTAB-1k, FGVC, few-shot learning, domain adaptation) showcasing its potential for resource-efficient fine-tuning. It also improves LoRA in text classification (GLUE) and mathematical reasoning (GSM-8K). The code is available at https://github.com/MaxwellYaoNi/PACE
comment: Accepted by NeurIPS 2024 as a spotlight
♻ ☆ Supervised Kernel Thinning NeurIPS 2024
The kernel thinning algorithm of Dwivedi & Mackey (2024) provides a better-than-i.i.d. compression of a generic set of points. By generating high-fidelity coresets of size significantly smaller than the input points, KT is known to speed up unsupervised tasks like Monte Carlo integration, uncertainty quantification, and non-parametric hypothesis testing, with minimal loss in statistical accuracy. In this work, we generalize the KT algorithm to speed up supervised learning problems involving kernel methods. Specifically, we combine two classical algorithms--Nadaraya-Watson (NW) regression or kernel smoothing, and kernel ridge regression (KRR)--with KT to provide a quadratic speed-up in both training and inference times. We show how distribution compression with KT in each setting reduces to constructing an appropriate kernel, and introduce the Kernel-Thinned NW and Kernel-Thinned KRR estimators. We prove that KT-based regression estimators enjoy significantly superior computational efficiency over the full-data estimators and improved statistical efficiency over i.i.d. subsampling of the training data. En route, we also provide a novel multiplicative error guarantee for compressing with KT. We validate our design choices with both simulations and real data experiments.
comment: Published at NeurIPS 2024
♻ ☆ Integrating Multi-Physics Simulations and Machine Learning to Define the Spatter Mechanism and Process Window in Laser Powder Bed Fusion
Laser powder bed fusion (LPBF) has shown promise for wide range of applications due to its ability to fabricate freeform geometries and generate a controlled microstructure. However, components generated by LPBF still possess sub-optimal mechanical properties due to the defects that are created during laser-material interactions. In this work, we investigate mechanism of spatter formation, using a high-fidelity modelling tool that was built to simulate the multi-physics phenomena in LPBF. The modelling tool have the capability to capture the 3D resolution of the meltpool and the spatter behavior. To understand spatter behavior and formation, we reveal its properties at ejection and evaluate its variation from the meltpool, the source where it is formed. The dataset of the spatter and the meltpool collected consist of 50 % spatter and 50 % melt pool samples, with features that include position components, velocity components, velocity magnitude, temperature, density and pressure. The relationship between the spatter and the meltpool were evaluated via correlation analysis and machine learning (ML) algorithms for classification tasks. Upon screening different ML algorithms on the dataset, a high accuracy was observed for all the ML models, with ExtraTrees having the highest at 96 % and KNN having the lowest at 94 %.
♻ ☆ Better by Default: Strong Pre-Tuned MLPs and Boosted Trees on Tabular Data NeurIPS 2024
For classification and regression on tabular data, the dominance of gradient-boosted decision trees (GBDTs) has recently been challenged by often much slower deep learning methods with extensive hyperparameter tuning. We address this discrepancy by introducing (a) RealMLP, an improved multilayer perceptron (MLP), and (b) strong meta-tuned default parameters for GBDTs and RealMLP. We tune RealMLP and the default parameters on a meta-train benchmark with 118 datasets and compare them to hyperparameter-optimized versions on a disjoint meta-test benchmark with 90 datasets, as well as the GBDT-friendly benchmark by Grinsztajn et al. (2022). Our benchmark results on medium-to-large tabular datasets (1K--500K samples) show that RealMLP offers a favorable time-accuracy tradeoff compared to other neural baselines and is competitive with GBDTs in terms of benchmark scores. Moreover, a combination of RealMLP and GBDTs with improved default parameters can achieve excellent results without hyperparameter tuning. Finally, we demonstrate that some of RealMLP's improvements can also considerably improve the performance of TabR with default parameters.
comment: NeurIPS 2024. Changes in v3: mention bug in XGBoost results, mention original name of he+5 method. Code is available at github.com/dholzmueller/pytabkit
♻ ☆ Ensemble sampling for linear bandits: small ensembles suffice
We provide the first useful and rigorous analysis of ensemble sampling for the stochastic linear bandit setting. In particular, we show that, under standard assumptions, for a $d$-dimensional stochastic linear bandit with an interaction horizon $T$, ensemble sampling with an ensemble of size of order $d \log T$ incurs regret at most of the order $(d \log T)^{5/2} \sqrt{T}$. Ours is the first result in any structured setting not to require the size of the ensemble to scale linearly with $T$ -- which defeats the purpose of ensemble sampling -- while obtaining near $\smash{\sqrt{T}}$ order regret. Our result is also the first to allow for infinite action sets.
♻ ☆ Inferring stochastic low-rank recurrent neural networks from neural data
A central aim in computational neuroscience is to relate the activity of large populations of neurons to an underlying dynamical system. Models of these neural dynamics should ideally be both interpretable and fit the observed data well. Low-rank recurrent neural networks (RNNs) exhibit such interpretability by having tractable dynamics. However, it is unclear how to best fit low-rank RNNs to data consisting of noisy observations of an underlying stochastic system. Here, we propose to fit stochastic low-rank RNNs with variational sequential Monte Carlo methods. We validate our method on several datasets consisting of both continuous and spiking neural data, where we obtain lower dimensional latent dynamics than current state of the art methods. Additionally, for low-rank models with piecewise linear nonlinearities, we show how to efficiently identify all fixed points in polynomial rather than exponential cost in the number of units, making analysis of the inferred dynamics tractable for large RNNs. Our method both elucidates the dynamical systems underlying experimental recordings and provides a generative model whose trajectories match observed variability.
♻ ☆ Taming the Long Tail in Human Mobility Prediction NeurIPS 2024
With the popularity of location-based services, human mobility prediction plays a key role in enhancing personalized navigation, optimizing recommendation systems, and facilitating urban mobility and planning. This involves predicting a user's next POI (point-of-interest) visit using their past visit history. However, the uneven distribution of visitations over time and space, namely the long-tail problem in spatial distribution, makes it difficult for AI models to predict those POIs that are less visited by humans. In light of this issue, we propose the Long-Tail Adjusted Next POI Prediction (LoTNext) framework for mobility prediction, combining a Long-Tailed Graph Adjustment module to reduce the impact of the long-tailed nodes in the user-POI interaction graph and a novel Long-Tailed Loss Adjustment module to adjust loss by logit score and sample weight adjustment strategy. Also, we employ the auxiliary prediction task to enhance generalization and accuracy. Our experiments with two real-world trajectory datasets demonstrate that LoTNext significantly surpasses existing state-of-the-art works.
comment: Accepted by NeurIPS 2024
♻ ☆ The Surprising Ineffectiveness of Pre-Trained Visual Representations for Model-Based Reinforcement Learning NeurIPS 2024
Visual Reinforcement Learning (RL) methods often require extensive amounts of data. As opposed to model-free RL, model-based RL (MBRL) offers a potential solution with efficient data utilization through planning. Additionally, RL lacks generalization capabilities for real-world tasks. Prior work has shown that incorporating pre-trained visual representations (PVRs) enhances sample efficiency and generalization. While PVRs have been extensively studied in the context of model-free RL, their potential in MBRL remains largely unexplored. In this paper, we benchmark a set of PVRs on challenging control tasks in a model-based RL setting. We investigate the data efficiency, generalization capabilities, and the impact of different properties of PVRs on the performance of model-based agents. Our results, perhaps surprisingly, reveal that for MBRL current PVRs are not more sample efficient than learning representations from scratch, and that they do not generalize better to out-of-distribution (OOD) settings. To explain this, we analyze the quality of the trained dynamics model. Furthermore, we show that data diversity and network architecture are the most important contributors to OOD generalization performance.
comment: Published at the 38th Conference on Neural Information Processing Systems (NeurIPS 2024). Project page: https://schneimo.com/pvr4mbrl/
♻ ☆ CGCOD: Class-Guided Camouflaged Object Detection
Camouflaged Object Detection (COD) aims to identify objects that blend seamlessly into their surroundings. The inherent visual complexity of camouflaged objects, including their low contrast with the background, diverse textures, and subtle appearance variations, often obscures semantic cues, making accurate segmentation highly challenging. Existing methods primarily rely on visual features, which are insufficient to handle the variability and intricacy of camouflaged objects, leading to unstable object perception and ambiguous segmentation results. To tackle these limitations, we introduce a novel task, class-guided camouflaged object detection (CGCOD), which extends traditional COD task by incorporating object-specific class knowledge to enhance detection robustness and accuracy. To facilitate this task, we present a new dataset, CamoClass, comprising real-world camouflaged objects with class annotations. Furthermore, we propose a multi-stage framework, CGNet, which incorporates a plug-and-play class prompt generator and a simple yet effective class-guided detector. This establishes a new paradigm for COD, bridging the gap between contextual understanding and class-guided detection. Extensive experimental results demonstrate the effectiveness of our flexible framework in improving the performance of proposed and existing detectors by leveraging class-level textual information.
♻ ☆ RoME: A Robust Mixed-Effects Bandit Algorithm for Optimizing Mobile Health Interventions
Mobile health leverages personalized and contextually tailored interventions optimized through bandit and reinforcement learning algorithms. In practice, however, challenges such as participant heterogeneity, nonstationarity, and nonlinear relationships hinder algorithm performance. We propose RoME, a Robust Mixed-Effects contextual bandit algorithm that simultaneously addresses these challenges via (1) modeling the differential reward with user- and time-specific random effects, (2) network cohesion penalties, and (3) debiased machine learning for flexible estimation of baseline rewards. We establish a high-probability regret bound that depends solely on the dimension of the differential-reward model, enabling us to achieve robust regret bounds even when the baseline reward is highly complex. We demonstrate the superior performance of the RoME algorithm in a simulation and two off-policy evaluation studies.
♻ ☆ Improved Algorithms for Contextual Dynamic Pricing
In contextual dynamic pricing, a seller sequentially prices goods based on contextual information. Buyers will purchase products only if the prices are below their valuations. The goal of the seller is to design a pricing strategy that collects as much revenue as possible. We focus on two different valuation models. The first assumes that valuations linearly depend on the context and are further distorted by noise. Under minor regularity assumptions, our algorithm achieves an optimal regret bound of $\tilde{\mathcal{O}}(T^{2/3})$, improving the existing results. The second model removes the linearity assumption, requiring only that the expected buyer valuation is $\beta$-H\"older in the context. For this model, our algorithm obtains a regret $\tilde{\mathcal{O}}(T^{d+2\beta/d+3\beta})$, where $d$ is the dimension of the context space.
♻ ☆ PRIMO: Private Regression in Multiple Outcomes
We introduce a new private regression setting we call Private Regression in Multiple Outcomes (PRIMO), inspired by the common situation where a data analyst wants to perform a set of $l$ regressions while preserving privacy, where the features $X$ are shared across all $l$ regressions, and each regression $i \in [l]$ has a different vector of outcomes $y_i$. Naively applying existing private linear regression techniques $l$ times leads to a $\sqrt{l}$ multiplicative increase in error over the standard linear regression setting. We apply a variety of techniques including sufficient statistics perturbation (SSP) and geometric projection-based methods to develop scalable algorithms that outperform this baseline across a range of parameter regimes. In particular, we obtain no dependence on l in the asymptotic error when $l$ is sufficiently large. Empirically, on the task of genomic risk prediction with multiple phenotypes we find that even for values of $l$ far smaller than the theory would predict, our projection-based method improves the accuracy relative to the variant that doesn't use the projection.
♻ ☆ Volterra Accentuated Non-Linear Dynamical Admittance (VANYA) to model Deforestation: An Exemplification from the Amazon Rainforest
Intelligent automation supports us against cyclones, droughts, and seismic events with recent technology advancements. Algorithmic learning has advanced fields like neuroscience, genetics, and human-computer interaction. Time-series data boosts progress. Challenges persist in adopting these approaches in traditional fields. Neural networks face comprehension and bias issues. AI's expansion across scientific areas is due to adaptable descriptors and combinatorial argumentation. This article focuses on modeling Forest loss using the VANYA Model, incorporating Prey Predator Dynamics. VANYA predicts forest cover, demonstrated on Amazon Rainforest data against other forecasters like Long Short-Term Memory, N-BEATS, RCN.
comment: The experimental data used in this article has given wrong practical interpretation. The data has to be updated to improve this
♻ ☆ Learning Optimal Tax Design in Nonatomic Congestion Games NeurIPS
In multiplayer games, self-interested behavior among the players can harm the social welfare. Tax mechanisms are a common method to alleviate this issue and induce socially optimal behavior. In this work, we take the initial step of learning the optimal tax that can maximize social welfare with limited feedback in congestion games. We propose a new type of feedback named \emph{equilibrium feedback}, where the tax designer can only observe the Nash equilibrium after deploying a tax plan. Existing algorithms are not applicable due to the exponentially large tax function space, nonexistence of the gradient, and nonconvexity of the objective. To tackle these challenges, we design a computationally efficient algorithm that leverages several novel components: (1) a piece-wise linear tax to approximate the optimal tax; (2) extra linear terms to guarantee a strongly convex potential function; (3) an efficient subroutine to find the exploratory tax that can provide critical information about the game. The algorithm can find an $\epsilon$-optimal tax with $O(\beta F^2/\epsilon)$ sample complexity, where $\beta$ is the smoothness of the cost function and $F$ is the number of facilities.
comment: 23 pages. Accepted by Conference on Neural Information Processing Systems (NeurIPS) 2024
♻ ☆ Evaluation of Artificial Intelligence Methods for Lead Time Prediction in Non-Cycled Areas of Automotive Production
The present study examines the effectiveness of applying Artificial Intelligence methods in an automotive production environment to predict unknown lead times in a non-cycle-controlled production area. Data structures are analyzed to identify contextual features and then preprocessed using one-hot encoding. Methods selection focuses on supervised machine learning techniques. In supervised learning methods, regression and classification methods are evaluated. Continuous regression based on target size distribution is not feasible. Classification methods analysis shows that Ensemble Learning and Support Vector Machines are the most suitable. Preliminary study results indicate that gradient boosting algorithms LightGBM, XGBoost, and CatBoost yield the best results. After further testing and extensive hyperparameter optimization, the final method choice is the LightGBM algorithm. Depending on feature availability and prediction interval granularity, relative prediction accuracies of up to 90% can be achieved. Further tests highlight the importance of periodic retraining of AI models to accurately represent complex production processes using the database. The research demonstrates that AI methods can be effectively applied to highly variable production data, adding business value by providing an additional metric for various control tasks while outperforming current non AI-based systems.
♻ ☆ Constrained Latent Action Policies for Model-Based Offline Reinforcement Learning NeurIPS 2024
In offline reinforcement learning, a policy is learned using a static dataset in the absence of costly feedback from the environment. In contrast to the online setting, only using static datasets poses additional challenges, such as policies generating out-of-distribution samples. Model-based offline reinforcement learning methods try to overcome these by learning a model of the underlying dynamics of the environment and using it to guide policy search. It is beneficial but, with limited datasets, errors in the model and the issue of value overestimation among out-of-distribution states can worsen performance. Current model-based methods apply some notion of conservatism to the Bellman update, often implemented using uncertainty estimation derived from model ensembles. In this paper, we propose Constrained Latent Action Policies (C-LAP) which learns a generative model of the joint distribution of observations and actions. We cast policy learning as a constrained objective to always stay within the support of the latent action distribution, and use the generative capabilities of the model to impose an implicit constraint on the generated actions. Thereby eliminating the need to use additional uncertainty penalties on the Bellman update and significantly decreasing the number of gradient steps required to learn a policy. We empirically evaluate C-LAP on the D4RL and V-D4RL benchmark, and show that C-LAP is competitive to state-of-the-art methods, especially outperforming on datasets with visual observations.
comment: 38th Conference on Neural Information Processing Systems (NeurIPS 2024)
♻ ☆ Metric Space Magnitude for Evaluating the Diversity of Latent Representations NeurIPS
The magnitude of a metric space is a novel invariant that provides a measure of the 'effective size' of a space across multiple scales, while also capturing numerous geometrical properties, such as curvature, density, or entropy. We develop a family of magnitude-based measures of the intrinsic diversity of latent representations, formalising a novel notion of dissimilarity between magnitude functions of finite metric spaces. Our measures are provably stable under perturbations of the data, can be efficiently calculated, and enable a rigorous multi-scale characterisation and comparison of latent representations. We show their utility and superior performance across different domains and tasks, including (i) the automated estimation of diversity, (ii) the detection of mode collapse, and (iii) the evaluation of generative models for text, image, and graph data.
comment: Accepted at the 38th Conference on Neural Information Processing Systems (NeurIPS) 2024. The code for computing magnitude is available at https://github.com/aidos-lab/magnipy
♻ ☆ Maximizing Uncertainty for Federated learning via Bayesian Optimisation-based Model Poisoning
As we transition from Narrow Artificial Intelligence towards Artificial Super Intelligence, users are increasingly concerned about their privacy and the trustworthiness of machine learning (ML) technology. A common denominator for the metrics of trustworthiness is the quantification of uncertainty inherent in DL algorithms, and specifically in the model parameters, input data, and model predictions. One of the common approaches to address privacy-related issues in DL is to adopt distributed learning such as federated learning (FL), where private raw data is not shared among users. Despite the privacy-preserving mechanisms in FL, it still faces challenges in trustworthiness. Specifically, the malicious users, during training, can systematically create malicious model parameters to compromise the models predictive and generative capabilities, resulting in high uncertainty about their reliability. To demonstrate malicious behaviour, we propose a novel model poisoning attack method named Delphi which aims to maximise the uncertainty of the global model output. We achieve this by taking advantage of the relationship between the uncertainty and the model parameters of the first hidden layer of the local model. Delphi employs two types of optimisation , Bayesian Optimisation and Least Squares Trust Region, to search for the optimal poisoned model parameters, named as Delphi-BO and Delphi-LSTR. We quantify the uncertainty using the KL Divergence to minimise the distance of the predictive probability distribution towards an uncertain distribution of model output. Furthermore, we establish a mathematical proof for the attack effectiveness demonstrated in FL. Numerical results demonstrate that Delphi-BO induces a higher amount of uncertainty than Delphi-LSTR highlighting vulnerability of FL systems to model poisoning attacks.
comment: 14 pages
♻ ☆ Applying the maximum entropy principle to neural networks enhances multi-species distribution models
The rapid expansion of citizen science initiatives has led to a significant growth of biodiversity databases, and particularly presence-only (PO) observations. PO data are invaluable for understanding species distributions and their dynamics, but their use in a Species Distribution Model (SDM) is curtailed by sampling biases and the lack of information on absences. Poisson point processes are widely used for SDMs, with Maxent being one of the most popular methods. Maxent maximises the entropy of a probability distribution across sites as a function of predefined transformations of variables, called features. In contrast, neural networks and deep learning have emerged as a promising technique for automatic feature extraction from complex input variables. Arbitrarily complex transformations of input variables can be learned from the data efficiently through backpropagation and stochastic gradient descent (SGD). In this paper, we propose DeepMaxent, which harnesses neural networks to automatically learn shared features among species, using the maximum entropy principle. To do so, it employs a normalised Poisson loss where for each species, presence probabilities across sites are modelled by a neural network. We evaluate DeepMaxent on a benchmark dataset known for its spatial sampling biases, using PO data for calibration and presence-absence (PA) data for validation across six regions with different biological groups and covariates. Our results indicate that DeepMaxent performs better than Maxent and other leading SDMs across all regions and taxonomic groups. The method performs particularly well in regions of uneven sampling, demonstrating substantial potential to increase SDM performances. In particular, our approach yields more accurate predictions than traditional single-species models, which opens up new possibilities for methodological enhancement.
comment: Submitted to Methods in Ecology and Evolution
♻ ☆ A Closer Look at Deep Learning Methods on Tabular Datasets
Tabular data is prevalent across diverse domains in machine learning. While classical methods like tree-based models have long been effective, Deep Neural Network (DNN)-based methods have recently demonstrated promising performance. However, the diverse characteristics of methods and the inherent heterogeneity of tabular datasets make understanding and interpreting tabular methods both challenging and prone to unstable observations. In this paper, we conduct in-depth evaluations and comprehensive analyses of tabular methods, with a particular focus on DNN-based models, using a benchmark of over 300 tabular datasets spanning a wide range of task types, sizes, and domains. First, we perform an extensive comparison of 32 state-of-the-art deep and tree-based methods, evaluating their average performance across multiple criteria. Although method ranks vary across datasets, we empirically find that top-performing methods tend to concentrate within a small subset of tabular models, regardless of the criteria used. Next, we investigate whether the training dynamics of deep tabular models can be predicted based on dataset properties. This approach not only offers insights into the behavior of deep tabular methods but also identifies a core set of "meta-features" that reflect dataset heterogeneity. The other subset includes datasets where method ranks are consistent with the overall benchmark, acting as a reliable probe for further tabular analysis.
♻ ☆ MambaLRP: Explaining Selective State Space Sequence Models
Recent sequence modeling approaches using selective state space sequence models, referred to as Mamba models, have seen a surge of interest. These models allow efficient processing of long sequences in linear time and are rapidly being adopted in a wide range of applications such as language modeling, demonstrating promising performance. To foster their reliable use in real-world scenarios, it is crucial to augment their transparency. Our work bridges this critical gap by bringing explainability, particularly Layer-wise Relevance Propagation (LRP), to the Mamba architecture. Guided by the axiom of relevance conservation, we identify specific components in the Mamba architecture, which cause unfaithful explanations. To remedy this issue, we propose MambaLRP, a novel algorithm within the LRP framework, which ensures a more stable and reliable relevance propagation through these components. Our proposed method is theoretically sound and excels in achieving state-of-the-art explanation performance across a diverse range of models and datasets. Moreover, MambaLRP facilitates a deeper inspection of Mamba architectures, uncovering various biases and evaluating their significance. It also enables the analysis of previous speculations regarding the long-range capabilities of Mamba models.
♻ ☆ Sparse Low-Ranked Self-Attention Transformer for Remaining Useful Lifetime Prediction of Optical Fiber Amplifiers
Optical fiber amplifiers are key elements in present optical networks. Failures of these components result in high financial loss of income of the network operator as the communication traffic over an affected link is interrupted. Applying Remaining useful lifetime (RUL) prediction in the context of Predictive Maintenance (PdM) to optical fiber amplifiers to predict upcoming system failures at an early stage, so that network outages can be minimized through planning of targeted maintenance actions, ensures reliability and safety. Optical fiber amplifier are complex systems, that work under various operating conditions, which makes correct forecasting a difficult task. Increased monitoring capabilities of systems results in datasets that facilitate the application of data-driven RUL prediction methods. Deep learning models in particular have shown good performance, but generalization based on comparatively small datasets for RUL prediction is difficult. In this paper, we propose Sparse Low-ranked self-Attention Transformer (SLAT) as a novel RUL prediction method. SLAT is based on an encoder-decoder architecture, wherein two parallel working encoders extract features for sensors and time steps. By utilizing the self-attention mechanism, long-term dependencies can be learned from long sequences. The implementation of sparsity in the attention matrix and a low-rank parametrization reduce overfitting and increase generalization. Experimental application to optical fiber amplifiers exemplified on EDFA, as well as a reference dataset from turbofan engines, shows that SLAT outperforms the state-of-the-art methods.
comment: 9 pages, 7 figures
♻ ☆ FADE: Towards Fairness-aware Augmentation for Domain Generalization via Classifier-Guided Score-based Diffusion Models
Fairness-aware domain generalization (FairDG) has emerged as a critical challenge for deploying trustworthy AI systems, particularly in scenarios involving distribution shifts. Traditional methods for addressing fairness have failed in domain generalization due to their lack of consideration for distribution shifts. Although disentanglement has been used to tackle FairDG, it is limited by its strong assumptions. To overcome these limitations, we propose Fairness-aware Classifier-Guided Score-based Diffusion Models (FADE) as a novel approach to effectively address the FairDG issue. Specifically, we first pre-train a score-based diffusion model (SDM) and two classifiers to equip the model with strong generalization capabilities across different domains. Then, we guide the SDM using these pre-trained classifiers to effectively eliminate sensitive information from the generated data. Finally, the generated fair data is used to train downstream classifiers, ensuring robust performance under new data distributions. Extensive experiments on three real-world datasets demonstrate that FADE not only enhances fairness but also improves accuracy in the presence of distribution shifts. Additionally, FADE outperforms existing methods in achieving the best accuracy-fairness trade-offs.
♻ ☆ Parallelizing Linear Transformers with the Delta Rule over Sequence Length
Transformers with linear attention (i.e., linear transformers) and state-space models have recently been suggested as a viable linear-time alternative to transformers with softmax attention. However, these models still underperform transformers especially on tasks that require in-context retrieval. While more expressive variants of linear transformers which replace the additive update in linear transformers with the delta rule (DeltaNet) have been found to be more effective at associative recall, existing algorithms for training such models do not parallelize over sequence length and are thus inefficient to train on modern hardware. This work describes a hardware-efficient algorithm for training linear transformers with the delta rule, which exploits a memory-efficient representation for computing products of Householder matrices. This algorithm allows us to scale up DeltaNet to standard language modeling settings. We train a 1.3B model for 100B tokens and find that it outperforms recent linear-time baselines such as Mamba and GLA in terms of perplexity and zero-shot performance on downstream tasks. We also experiment with two hybrid models which combine DeltaNet layers with (1) sliding-window attention layers every other layer or (2) two global attention layers, and find that these hybrids outperform strong transformer baselines.
comment: Final camera ready
♻ ☆ RoHan: Robust Hand Detection in Operation Room
Hand-specific localization has garnered significant interest within the computer vision community. Although there are numerous datasets with hand annotations from various angles and settings, domain transfer techniques frequently struggle in surgical environments. This is mainly due to the limited availability of gloved hand instances and the unique challenges of operating rooms (ORs). Thus, hand-detection models tailored to OR settings require extensive training and expensive annotation processes. To overcome these challenges, we present "RoHan" - a novel approach for robust hand detection in the OR, leveraging advanced semi-supervised domain adaptation techniques to tackle the challenges of varying recording conditions, diverse glove colors, and occlusions common in surgical settings. Our methodology encompasses two main stages: (1) data augmentation strategy that utilizes "Artificial Gloves," a method for augmenting publicly available hand datasets with synthetic images of hands-wearing gloves; (2) semi-supervised domain adaptation pipeline that improves detection performance in real-world OR settings through iterative prediction refinement and efficient frame filtering. We evaluate our method using two datasets: simulated enterotomy repair and saphenous vein graft harvesting. "RoHan" substantially reduces the need for extensive labeling and model training, paving the way for the practical implementation of hand detection technologies in medical settings.
comment: 12 pages
♻ ☆ Constructing Confidence Intervals for 'the' Generalization Error -- a Comprehensive Benchmark Study
When assessing the quality of prediction models in machine learning, confidence intervals (CIs) for the generalization error, which measures predictive performance, are a crucial tool. Luckily, there exist many methods for computing such CIs and new promising approaches are continuously being proposed. Typically, these methods combine various resampling procedures, most popular among them cross-validation and bootstrapping, with different variance estimation techniques. Unfortunately, however, there is currently no consensus on when any of these combinations may be most reliably employed and how they generally compare. In this work, we conduct a large-scale study comparing CIs for the generalization error, the first one of such size, where we empirically evaluate 13 different CI methods on a total of 19 tabular regression and classification problems, using seven different inducers and a total of eight loss functions. We give an overview of the methodological foundations and inherent challenges of constructing CIs for the generalization error and provide a concise review of all 13 methods in a unified framework. Finally, the CI methods are evaluated in terms of their relative coverage frequency, width, and runtime. Based on these findings, we can identify a subset of methods that we would recommend. We also publish the datasets as a benchmarking suite on OpenML and our code on GitHub to serve as a basis for further studies.
♻ ☆ Extended convexity and smoothness and their applications in deep learning
This paper introduces an optimization framework aimed at providing a theoretical foundation for a class of composite optimization problems, particularly those encountered in deep learning. In this framework, we introduce $\mathcal{H}(\phi)$-convexity and $\mathcal{H}(\Phi)$-smoothness to generalize the existing concepts of Lipschitz smoothness and strong convexity. Furthermore, we analyze and establish the convergence of both gradient descent and stochastic gradient descent methods for objective functions that are $\mathcal{H}(\Phi)$-smooth. We prove that the optimal convergence rates of these methods depend solely on the homogeneous degree of $\Phi$. Based on these findings, we construct two types of non-convex and non-smooth optimization problems: deterministic composite and stochastic composite optimization problems, which encompass the majority of optimization problems in deep learning. To address these problems, we develop the gradient structure control algorithm and prove that it can locate approximate global optima. This marks a significant departure from traditional non-convex analysis framework, which typically settle for stationary points. Therefore, with the introduction of $\mathcal{H}(\phi)$-convexity and $\mathcal{H}(\Phi)$-smoothness, along with the GSC algorithm, the non-convex optimization mechanisms in deep learning can be theoretically explained and supported. Finally, the effectiveness of the proposed framework is substantiated through empirical experimentation.
♻ ☆ Diffusion-based Unsupervised Audio-visual Speech Enhancement
This paper proposes a new unsupervised audio-visual speech enhancement (AVSE) approach that combines a diffusion-based audio-visual speech generative model with a non-negative matrix factorization (NMF) noise model. First, the diffusion model is pre-trained on clean speech conditioned on corresponding video data to simulate the speech generative distribution. This pre-trained model is then paired with the NMF-based noise model to estimate clean speech iteratively. Specifically, a diffusion-based posterior sampling approach is implemented within the reverse diffusion process, where after each iteration, a speech estimate is obtained and used to update the noise parameters. Experimental results confirm that the proposed AVSE approach not only outperforms its audio-only counterpart but also generalizes better than a recent supervised-generative AVSE method. Additionally, the new inference algorithm offers a better balance between inference speed and performance compared to the previous diffusion-based method. Code and demo available at: https://jeaneudesayilo.github.io/fast_UdiffSE
♻ ☆ Interpreting Equivariant Representations ICML 2024
Latent representations are used extensively for downstream tasks, such as visualization, interpolation or feature extraction of deep learning models. Invariant and equivariant neural networks are powerful and well-established models for enforcing inductive biases. In this paper, we demonstrate that the inductive bias imposed on the by an equivariant model must also be taken into account when using latent representations. We show how not accounting for the inductive biases leads to decreased performance on downstream tasks, and vice versa, how accounting for inductive biases can be done effectively by using an invariant projection of the latent representations. We propose principles for how to choose such a projection, and show the impact of using these principles in two common examples: First, we study a permutation equivariant variational auto-encoder trained for molecule graph generation; here we show that invariant projections can be designed that incur no loss of information in the resulting invariant representation. Next, we study a rotation-equivariant representation used for image classification. Here, we illustrate how random invariant projections can be used to obtain an invariant representation with a high degree of retained information. In both cases, the analysis of invariant latent representations proves superior to their equivariant counterparts. Finally, we illustrate that the phenomena documented here for equivariant neural networks have counterparts in standard neural networks where invariance is encouraged via augmentation. Thus, while these ambiguities may be known by experienced developers of equivariant models, we make both the knowledge as well as effective tools to handle the ambiguities available to the broader community.
comment: This paper was updated to reflect the version accepted to ICML 2024
♻ ☆ A Unified Confidence Sequence for Generalized Linear Models, with Applications to Bandits NeurIPS 2024
We present a unified likelihood ratio-based confidence sequence (CS) for any (self-concordant) generalized linear model (GLM) that is guaranteed to be convex and numerically tight. We show that this is on par or improves upon known CSs for various GLMs, including Gaussian, Bernoulli, and Poisson. In particular, for the first time, our CS for Bernoulli has a $\mathrm{poly}(S)$-free radius where $S$ is the norm of the unknown parameter. Our first technical novelty is its derivation, which utilizes a time-uniform PAC-Bayesian bound with a uniform prior/posterior, despite the latter being a rather unpopular choice for deriving CSs. As a direct application of our new CS, we propose a simple and natural optimistic algorithm called OFUGLB, applicable to any generalized linear bandits (GLB; Filippi et al. (2010)). Our analysis shows that the celebrated optimistic approach simultaneously attains state-of-the-art regrets for various self-concordant (not necessarily bounded) GLBs, and even $\mathrm{poly}(S)$-free for bounded GLBs, including logistic bandits. The regret analysis, our second technical novelty, follows from combining our new CS with a new proof technique that completely avoids the previously widely used self-concordant control lemma (Faury et al., 2020, Lemma 9). Numerically, OFUGLB outperforms or is at par with prior algorithms for logistic bandits.
comment: 39 pages, 2 figures, 2 tables; Accepted to the 38th Conference on Neural Information Processing Systems (NeurIPS 2024) (ver3: minor revisions, code refactoring; ver2: major revision, including new experiments, reorganization, fixing typos in the proofs of ver1, etc)
♻ ☆ SupplyGraph: A Benchmark Dataset for Supply Chain Planning using Graph Neural Networks AAAI 2024
Graph Neural Networks (GNNs) have gained traction across different domains such as transportation, bio-informatics, language processing, and computer vision. However, there is a noticeable absence of research on applying GNNs to supply chain networks. Supply chain networks are inherently graph-like in structure, making them prime candidates for applying GNN methodologies. This opens up a world of possibilities for optimizing, predicting, and solving even the most complex supply chain problems. A major setback in this approach lies in the absence of real-world benchmark datasets to facilitate the research and resolution of supply chain problems using GNNs. To address the issue, we present a real-world benchmark dataset for temporal tasks, obtained from one of the leading FMCG companies in Bangladesh, focusing on supply chain planning for production purposes. The dataset includes temporal data as node features to enable sales predictions, production planning, and the identification of factory issues. By utilizing this dataset, researchers can employ GNNs to address numerous supply chain problems, thereby advancing the field of supply chain analytics and planning. Source: https://github.com/CIOL-SUST/SupplyGraph
comment: Accepted to 4th workshop on Graphs and more Complex structures for Learning and Reasoning, colocated with AAAI 2024
♻ ☆ Get Rid of Isolation: A Continuous Multi-task Spatio-Temporal Learning Framework NeurIPS 2024
Spatiotemporal learning has become a pivotal technique to enable urban intelligence. Traditional spatiotemporal models mostly focus on a specific task by assuming a same distribution between training and testing sets. However, given that urban systems are usually dynamic, multi-sourced with imbalanced data distributions, current specific task-specific models fail to generalize to new urban conditions and adapt to new domains without explicitly modeling interdependencies across various dimensions and types of urban data. To this end, we argue that there is an essential to propose a Continuous Multi-task Spatio-Temporal learning framework (CMuST) to empower collective urban intelligence, which reforms the urban spatiotemporal learning from single-domain to cooperatively multi-dimensional and multi-task learning. Specifically, CMuST proposes a new multi-dimensional spatiotemporal interaction network (MSTI) to allow cross-interactions between context and main observations as well as self-interactions within spatial and temporal aspects to be exposed, which is also the core for capturing task-level commonality and personalization. To ensure continuous task learning, a novel Rolling Adaptation training scheme (RoAda) is devised, which not only preserves task uniqueness by constructing data summarization-driven task prompts, but also harnesses correlated patterns among tasks by iterative model behavior modeling. We further establish a benchmark of three cities for multi-task spatiotemporal learning, and empirically demonstrate the superiority of CMuST via extensive evaluations on these datasets. The impressive improvements on both few-shot streaming data and new domain tasks against existing SOAT methods are achieved. Code is available at https://github.com/DILab-USTCSZ/CMuST.
comment: Accepted by NeurIPS 2024
♻ ☆ Fully Distributed, Flexible Compositional Visual Representations via Soft Tensor Products
Since the inception of the classicalist vs. connectionist debate, it has been argued that the ability to systematically combine symbol-like entities into compositional representations is crucial for human intelligence. In connectionist systems, the field of disentanglement has gained prominence for its ability to produce explicitly compositional representations; however, it relies on a fundamentally symbolic, concatenative representation of compositional structure that clashes with the continuous, distributed foundations of deep learning. To resolve this tension, we extend Smolensky's Tensor Product Representation (TPR) and introduce Soft TPR, a representational form that encodes compositional structure in an inherently distributed, flexible manner, along with Soft TPR Autoencoder, a theoretically-principled architecture designed specifically to learn Soft TPRs. Comprehensive evaluations in the visual representation learning domain demonstrate that the Soft TPR framework consistently outperforms conventional disentanglement alternatives -- achieving state-of-the-art disentanglement, boosting representation learner convergence, and delivering superior sample efficiency and low-sample regime performance in downstream tasks. These findings highlight the promise of a distributed and flexible approach to representing compositional structure by potentially enhancing alignment with the core principles of deep learning over the conventional symbolic approach.
comment: Accepted to Neurips 2024. 10 pages + supplementary
♻ ☆ SelectIT: Selective Instruction Tuning for LLMs via Uncertainty-Aware Self-Reflection NeurIPS 2024
Instruction tuning (IT) is crucial to tailoring large language models (LLMs) towards human-centric interactions. Recent advancements have shown that the careful selection of a small, high-quality subset of IT data can significantly enhance the performance of LLMs. Despite this, common approaches often rely on additional models or data, which increases costs and limits widespread adoption. In this work, we propose a novel approach, termed SelectIT, that capitalizes on the foundational capabilities of the LLM itself. Specifically, we exploit the intrinsic uncertainty present in LLMs to more effectively select high-quality IT data, without the need for extra resources. Furthermore, we introduce a curated IT dataset, the Selective Alpaca, created by applying SelectIT to the Alpaca-GPT4 dataset. Empirical results demonstrate that IT using Selective Alpaca leads to substantial model ability enhancement. The robustness of SelectIT has also been corroborated in various foundation models and domain-specific tasks. Our findings suggest that longer and more computationally intensive IT data may serve as superior sources of IT, offering valuable insights for future research in this area. Data, code, and scripts are freely available at https://github.com/Blue-Raincoat/SelectIT.
comment: Accepted to NeurIPS 2024
♻ ☆ An Accelerated Algorithm for Stochastic Bilevel Optimization under Unbounded Smoothness NeurIPS 2024
This paper investigates a class of stochastic bilevel optimization problems where the upper-level function is nonconvex with potentially unbounded smoothness and the lower-level problem is strongly convex. These problems have significant applications in sequential data learning, such as text classification using recurrent neural networks. The unbounded smoothness is characterized by the smoothness constant of the upper-level function scaling linearly with the gradient norm, lacking a uniform upper bound. Existing state-of-the-art algorithms require $\widetilde{O}(1/\epsilon^4)$ oracle calls of stochastic gradient or Hessian/Jacobian-vector product to find an $\epsilon$-stationary point. However, it remains unclear if we can further improve the convergence rate when the assumptions for the function in the population level also hold for each random realization almost surely. To address this issue, we propose a new Accelerated Bilevel Optimization algorithm named AccBO. The algorithm updates the upper-level variable by normalized stochastic gradient descent with recursive momentum and the lower-level variable by the stochastic Nesterov accelerated gradient descent algorithm with averaging. We prove that our algorithm achieves an oracle complexity of $\widetilde{O}(1/\epsilon^3)$ to find an $\epsilon$-stationary point, when the lower-level stochastic gradient's variance is $O(\epsilon)$. Our proof relies on a novel lemma characterizing the dynamics of stochastic Nesterov accelerated gradient descent algorithm under distribution drift with high probability for the lower-level variable, which is of independent interest and also plays a crucial role in analyzing the hypergradient estimation error over time. Experimental results on various tasks confirm that our proposed algorithm achieves the predicted theoretical acceleration and significantly outperforms baselines in bilevel optimization.
comment: Accepted by NeurIPS 2024. The code is available at https://github.com/MingruiLiu-ML-Lab/Accelerated-Bilevel-Optimization-Unbounded-Smoothness
♻ ☆ Making AI Less "Thirsty": Uncovering and Addressing the Secret Water Footprint of AI Models
The growing carbon footprint of artificial intelligence (AI) has been undergoing public scrutiny. Nonetheless, the equally important water (withdrawal and consumption) footprint of AI has largely remained under the radar. For example, training the GPT-3 language model in Microsoft's state-of-the-art U.S. data centers can directly evaporate 700,000 liters of clean freshwater, but such information has been kept a secret. More critically, the global AI demand is projected to account for 4.2-6.6 billion cubic meters of water withdrawal in 2027, which is more than the total annual water withdrawal of 4-6 Denmark or half of the United Kingdom. This is concerning, as freshwater scarcity has become one of the most pressing challenges. To respond to the global water challenges, AI can, and also must, take social responsibility and lead by example by addressing its own water footprint. In this paper, we provide a principled methodology to estimate the water footprint of AI, and also discuss the unique spatial-temporal diversities of AI's runtime water efficiency. Finally, we highlight the necessity of holistically addressing water footprint along with carbon footprint to enable truly sustainable AI.
comment: Accepted by Communications of the ACM. Source codes available at: https://github.com/Ren-Research/Making-AI-Less-Thirsty
♻ ☆ MEMO: Fine-grained Tensor Management For Ultra-long Context LLM Training
Nowadays, Large Language Models (LLMs) have been trained using extended context lengths to foster more creative applications. However, long context training poses great challenges considering the constraint of GPU memory. It not only leads to substantial activation memory consumption during training, but also incurs considerable memory fragmentation. To facilitate long context training, existing frameworks have adopted strategies such as recomputation and various forms of parallelisms. Nevertheless, these techniques rely on redundant computation or extensive communication, resulting in low Model FLOPS Utilization (MFU). In this paper, we propose MEMO, a novel LLM training framework designed for fine-grained activation memory management. Given the quadratic scaling of computation and linear scaling of memory with sequence lengths when using FlashAttention, we offload memory-consuming activations to CPU memory after each layer's forward pass and fetch them during the backward pass. To maximize the swapping of activations without hindering computation, and to avoid exhausting limited CPU memory, we implement a token-wise activation recomputation and swapping mechanism. Furthermore, we tackle the memory fragmentation issue by employing a bi-level Mixed Integer Programming (MIP) approach, optimizing memory reuse across transformer layers. Empirical results demonstrate that MEMO achieves an average of 1.97x and 1.80x MFU compared to Megatron-LM and DeepSpeed, respectively. This improvement is attributed to MEMO's ability to minimize memory fragmentation, reduce recomputation and intensive communication, and circumvent the delays associated with the memory reorganization process due to fragmentation. By leveraging fine-grained activation memory management, MEMO facilitates efficient training of 7B LLM with 1 million sequence length on just 8 A800 GPUs, achieving an MFU of 52.30%.
♻ ☆ Dynamic Localisation of Spatial-Temporal Graph Neural Network KDD'25
Spatial-temporal data, fundamental to many intelligent applications, reveals dependencies indicating causal links between present measurements at specific locations and historical data at the same or other locations. Within this context, adaptive spatial-temporal graph neural networks (ASTGNNs) have emerged as valuable tools for modelling these dependencies, especially through a data-driven approach rather than pre-defined spatial graphs. While this approach offers higher accuracy, it presents increased computational demands. Addressing this challenge, this paper delves into the concept of localisation within ASTGNNs, introducing an innovative perspective that spatial dependencies should be dynamically evolving over time. We introduce \textit{DynAGS}, a localised ASTGNN framework aimed at maximising efficiency and accuracy in distributed deployment. This framework integrates dynamic localisation, time-evolving spatial graphs, and personalised localisation, all orchestrated around the Dynamic Graph Generator, a light-weighted central module leveraging cross attention. The central module can integrate historical information in a node-independent manner to enhance the feature representation of nodes at the current moment. This improved feature representation is then used to generate a dynamic sparse graph without the need for costly data exchanges, and it supports personalised localisation. Performance assessments across two core ASTGNN architectures and nine real-world datasets from various applications reveal that \textit{DynAGS} outshines current benchmarks, underscoring that the dynamic modelling of spatial dependencies can drastically improve model expressibility, flexibility, and system efficiency, especially in distributed settings.
comment: This paper was accepted by KDD'25
♻ ☆ OminiControl: Minimal and Universal Control for Diffusion Transformer
In this paper, we introduce OminiControl, a highly versatile and parameter-efficient framework that integrates image conditions into pre-trained Diffusion Transformer (DiT) models. At its core, OminiControl leverages a parameter reuse mechanism, enabling the DiT to encode image conditions using itself as a powerful backbone and process them with its flexible multi-modal attention processors. Unlike existing methods, which rely heavily on additional encoder modules with complex architectures, OminiControl (1) effectively and efficiently incorporates injected image conditions with only ~0.1% additional parameters, and (2) addresses a wide range of image conditioning tasks in a unified manner, including subject-driven generation and spatially-aligned conditions such as edges, depth, and more. Remarkably, these capabilities are achieved by training on images generated by the DiT itself, which is particularly beneficial for subject-driven generation. Extensive evaluations demonstrate that OminiControl outperforms existing UNet-based and DiT-adapted models in both subject-driven and spatially-aligned conditional generation. Additionally, we release our training dataset, Subjects200K, a diverse collection of over 200,000 identity-consistent images, along with an efficient data synthesis pipeline to advance research in subject-consistent generation.
♻ ☆ Diffusion Models as Network Optimizers: Explorations and Analysis
Network optimization is a fundamental challenge in the Internet of Things (IoT) network, often characterized by complex features that make it difficult to solve these problems. Recently, generative diffusion models (GDMs) have emerged as a promising new approach to network optimization, with the potential to directly address these optimization problems. However, the application of GDMs in this field is still in its early stages, and there is a noticeable lack of theoretical research and empirical findings. In this study, we first explore the intrinsic characteristics of generative models. Next, we provide a concise theoretical proof and intuitive demonstration of the advantages of generative models over discriminative models in network optimization. Based on this exploration, we implement GDMs as optimizers aimed at learning high-quality solution distributions for given inputs, sampling from these distributions during inference to approximate or achieve optimal solutions. Specifically, we utilize denoising diffusion probabilistic models (DDPMs) and employ a classifier-free guidance mechanism to manage conditional guidance based on input parameters. We conduct extensive experiments across three challenging network optimization problems. By investigating various model configurations and the principles of GDMs as optimizers, we demonstrate the ability to overcome prediction errors and validate the convergence of generated solutions to optimal solutions. We provide code and data at https://github.com/qiyu3816/DiffSG.
♻ ☆ CrossFi: A Cross Domain Wi-Fi Sensing Framework Based on Siamese Network
In recent years, Wi-Fi sensing has garnered significant attention due to its numerous benefits, such as privacy protection, low cost, and penetration ability. Extensive research has been conducted in this field, focusing on areas such as gesture recognition, people identification, and fall detection. However, many data-driven methods encounter challenges related to domain shift, where the model fails to perform well in environments different from the training data. One major factor contributing to this issue is the limited availability of Wi-Fi sensing datasets, which makes models learn excessive irrelevant information and over-fit to the training set. Unfortunately, collecting large-scale Wi-Fi sensing datasets across diverse scenarios is a challenging task. To address this problem, we propose CrossFi, a siamese network-based approach that excels in both in-domain scenario and cross-domain scenario, including few-shot, zero-shot scenarios, and even works in few-shot new-class scenario where testing set contains new categories. The core component of CrossFi is a sample-similarity calculation network called CSi-Net, which improves the structure of the siamese network by using an attention mechanism to capture similarity information, instead of simply calculating the distance or cosine similarity. Based on it, we develop an extra Weight-Net that can generate a template for each class, so that our CrossFi can work in different scenarios. Experimental results demonstrate that our CrossFi achieves state-of-the-art performance across various scenarios. In gesture recognition task, our CrossFi achieves an accuracy of 98.17% in in-domain scenario, 91.72% in one-shot cross-domain scenario, 64.81% in zero-shot cross-domain scenario, and 84.75% in one-shot new-class scenario. The code for our model is publicly available at https://github.com/RS2002/CrossFi.
♻ ☆ The Silent Majority: Demystifying Memorization Effect in the Presence of Spurious Correlations
Machine learning models often rely on simple spurious features -- patterns in training data that correlate with targets but are not causally related to them, like image backgrounds in foreground classification. This reliance typically leads to imbalanced test performance across minority and majority groups. In this work, we take a closer look at the fundamental cause of such imbalanced performance through the lens of memorization, which refers to the ability to predict accurately on \textit{atypical} examples (minority groups) in the training set but failing in achieving the same accuracy in the testing set. This paper systematically shows the ubiquitous existence of spurious features in a small set of neurons within the network, providing the first-ever evidence that memorization may contribute to imbalanced group performance. Through three experimental sources of converging empirical evidence, we find the property of a small subset of neurons or channels in memorizing minority group information. Inspired by these findings, we articulate the hypothesis: the imbalanced group performance is a byproduct of ``noisy'' spurious memorization confined to a small set of neurons. To further substantiate this hypothesis, we show that eliminating these unnecessary spurious memorization patterns via a novel framework during training can significantly affect the model performance on minority groups. Our experimental results across various architectures and benchmarks offer new insights on how neural networks encode core and spurious knowledge, laying the groundwork for future research in demystifying robustness to spurious correlation.
♻ ☆ Zero-shot Video Restoration and Enhancement Using Pre-Trained Image Diffusion Model AAAI 2025
Diffusion-based zero-shot image restoration and enhancement models have achieved great success in various tasks of image restoration and enhancement. However, directly applying them to video restoration and enhancement results in severe temporal flickering artifacts. In this paper, we propose the first framework for zero-shot video restoration and enhancement based on the pre-trained image diffusion model. By replacing the spatial self-attention layer with the proposed short-long-range (SLR) temporal attention layer, the pre-trained image diffusion model can take advantage of the temporal correlation between frames. We further propose temporal consistency guidance, spatial-temporal noise sharing, and an early stopping sampling strategy to improve temporally consistent sampling. Our method is a plug-and-play module that can be inserted into any diffusion-based image restoration or enhancement methods to further improve their performance. Experimental results demonstrate the superiority of our proposed method. Our code is available at https://github.com/cao-cong/ZVRD.
comment: Accepted by AAAI 2025
♻ ☆ Machine unlearning through fine-grained model parameters perturbation
Machine unlearning techniques, which involve retracting data records and reducing influence of said data on trained models, help with the user privacy protection objective but incur significant computational costs. Weight perturbation-based unlearning is a general approach, but it typically involves globally modifying the parameters. We propose fine-grained Top-K and Random-k parameters perturbed inexact machine unlearning strategies that address the privacy needs while keeping the computational costs tractable. In order to demonstrate the efficacy of our strategies we also tackle the challenge of evaluating the effectiveness of machine unlearning by considering the model's generalization performance across both unlearning and remaining data. To better assess the unlearning effect and model generalization, we propose novel metrics, namely, the forgetting rate and memory retention rate. However, for inexact machine unlearning, current metrics are inadequate in quantifying the degree of forgetting that occurs after unlearning strategies are applied. To address this, we introduce SPD-GAN, which subtly perturbs the distribution of data targeted for unlearning. Then, we evaluate the degree of unlearning by measuring the performance difference of the models on the perturbed unlearning data before and after the unlearning process. By implementing these innovative techniques and metrics, we achieve computationally efficacious privacy protection in machine learning applications without significant sacrifice of model performance. Furthermore, this approach provides a novel method for evaluating the degree of unlearning.
♻ ☆ Clarify Confused Nodes via Separated Learning
Graph neural networks (GNNs) have achieved remarkable advances in graph-oriented tasks. However, real-world graphs invariably contain a certain proportion of heterophilous nodes, challenging the homophily assumption of traditional GNNs and hindering their performance. Most existing studies continue to design generic models with shared weights between heterophilous and homophilous nodes. Despite the incorporation of high-order messages or multi-channel architectures, these efforts often fall short. A minority of studies attempt to train different node groups separately but suffer from inappropriate separation metrics and low efficiency. In this paper, we first propose a new metric, termed Neighborhood Confusion (NC), to facilitate a more reliable separation of nodes. We observe that node groups with different levels of NC values exhibit certain differences in intra-group accuracy and visualized embeddings. These pave the way for Neighborhood Confusion-guided Graph Convolutional Network (NCGCN), in which nodes are grouped by their NC values and accept intra-group weight sharing and message passing. Extensive experiments on both homophilous and heterophilous benchmarks demonstrate that our framework can effectively separate nodes and yield significant performance improvement compared to the latest methods. The source code will be available in https://github.com/GISec-Team/NCGNN.
comment: Accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence
♻ ☆ STORM: A Spatio-Temporal Factor Model Based on Dual Vector Quantized Variational Autoencoders for Financial Trading
In financial trading, factor models are widely used to price assets and capture excess returns from mispricing. Recently, we have witnessed the rise of variational autoencoder-based latent factor models, which learn latent factors self-adaptively. While these models focus on modeling overall market conditions, they often fail to effectively capture the temporal patterns of individual stocks. Additionally, representing multiple factors as single values simplifies the model but limits its ability to capture complex relationships and dependencies. As a result, the learned factors are of low quality and lack diversity, reducing their effectiveness and robustness across different trading periods. To address these issues, we propose a Spatio-Temporal factOR Model based on dual vector quantized variational autoencoders, named STORM, which extracts features of stocks from temporal and spatial perspectives, then fuses and aligns these features at the fine-grained and semantic level, and represents the factors as multi-dimensional embeddings. The discrete codebooks cluster similar factor embeddings, ensuring orthogonality and diversity, which helps distinguish between different factors and enables factor selection in financial trading. To show the performance of the proposed factor model, we apply it to two downstream experiments: portfolio management on two stock datasets and individual trading tasks on six specific stocks. The extensive experiments demonstrate STORM's flexibility in adapting to downstream tasks and superior performance over baseline models.
♻ ☆ Dual Cone Gradient Descent for Training Physics-Informed Neural Networks
Physics-informed neural networks (PINNs) have emerged as a prominent approach for solving partial differential equations (PDEs) by minimizing a combined loss function that incorporates both boundary loss and PDE residual loss. Despite their remarkable empirical performance in various scientific computing tasks, PINNs often fail to generate reasonable solutions, and such pathological behaviors remain difficult to explain and resolve. In this paper, we identify that PINNs can be adversely trained when gradients of each loss function exhibit a significant imbalance in their magnitudes and present a negative inner product value. To address these issues, we propose a novel optimization framework, Dual Cone Gradient Descent (DCGD), which adjusts the direction of the updated gradient to ensure it falls within a dual cone region. This region is defined as a set of vectors where the inner products with both the gradients of the PDE residual loss and the boundary loss are non-negative. Theoretically, we analyze the convergence properties of DCGD algorithms in a non-convex setting. On a variety of benchmark equations, we demonstrate that DCGD outperforms other optimization algorithms in terms of various evaluation metrics. In particular, DCGD achieves superior predictive accuracy and enhances the stability of training for failure modes of PINNs and complex PDEs, compared to existing optimally tuned models. Moreover, DCGD can be further improved by combining it with popular strategies for PINNs, including learning rate annealing and the Neural Tangent Kernel (NTK).
comment: The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024
♻ ☆ Conformal-in-the-Loop for Learning with Imbalanced Noisy Data
Class imbalance and label noise are pervasive in large-scale datasets, yet much of machine learning research assumes well-labeled, balanced data, which rarely reflects real world conditions. Existing approaches typically address either label noise or class imbalance in isolation, leading to suboptimal results when both issues coexist. In this work, we propose Conformal-in-the-Loop (CitL), a novel training framework that addresses both challenges with a conformal prediction-based approach. CitL evaluates sample uncertainty to adjust weights and prune unreliable examples, enhancing model resilience and accuracy with minimal computational cost. Our extensive experiments include a detailed analysis showing how CitL effectively emphasizes impactful data in noisy, imbalanced datasets. Our results show that CitL consistently boosts model performance, achieving up to a 6.1% increase in classification accuracy and a 5.0 mIoU improvement in segmentation. Our code is publicly available: CitL.
comment: Under Review
♻ ☆ EdgeSight: Enabling Modeless and Cost-Efficient Inference at the Edge
Traditional ML inference is evolving toward modeless inference, which abstracts the complexity of model selection from users, allowing the system to automatically choose the most appropriate model for each request based on accuracy and resource requirements. While prior studies have focused on modeless inference within data centers, this paper tackles the pressing need for cost-efficient modeless inference at the edge -- particularly within its unique constraints of limited device memory, volatile network conditions, and restricted power consumption. To overcome these challenges, we propose EdgeSight, a system that provides cost-efficient EdgeSight serving for diverse DNNs at the edge. EdgeSight employs an edge-data center (edge-DC) architecture, utilizing confidence scaling to reduce the number of model options while meeting diverse accuracy requirements. Additionally, it supports lossy inference in volatile network environments. Our experimental results show that EdgeSight outperforms existing systems by up to 1.6x in P99 latency for modeless services. Furthermore, our FPGA prototype demonstrates similar performance at certain accuracy levels, with a power consumption reduction of up to 3.34x.
comment: 12 pages
♻ ☆ Natural Language Outlines for Code: Literate Programming in the LLM Era
We propose using natural language outlines as a novel modality and interaction surface for providing AI assistance to developers throughout the software development process. An NL outline for a code function comprises multiple statements written in concise prose, which partition the code and summarize its main ideas in the style of literate programming. Crucially, we find that modern LLMs can generate accurate and high-quality NL outlines in practice. Moreover, NL outlines enable a bidirectional sync between code and NL, allowing changes in one to be automatically reflected in the other. We discuss many use cases for NL outlines: they can accelerate understanding and navigation of code and diffs, simplify code maintenance, augment code search, steer code generation, and more. We then propose and compare multiple LLM prompting techniques for generating outlines and ask professional developers to judge outline quality. Finally, we present two case studies applying NL outlines toward code review and malware detection.
♻ ☆ Continual Diffuser (CoD): Mastering Continual Offline Reinforcement Learning with Experience Rehearsal
Artificial neural networks, especially recent diffusion-based models, have shown remarkable superiority in gaming, control, and QA systems, where the training tasks' datasets are usually static. However, in real-world applications, such as robotic control of reinforcement learning (RL), the tasks are changing, and new tasks arise in a sequential order. This situation poses the new challenge of plasticity-stability trade-off for training an agent who can adapt to task changes and retain acquired knowledge. In view of this, we propose a rehearsal-based continual diffusion model, called Continual Diffuser (CoD), to endow the diffuser with the capabilities of quick adaptation (plasticity) and lasting retention (stability). Specifically, we first construct an offline benchmark that contains 90 tasks from multiple domains. Then, we train the CoD on each task with sequential modeling and conditional generation for making decisions. Next, we preserve a small portion of previous datasets as the rehearsal buffer and replay it to retain the acquired knowledge. Extensive experiments on a series of tasks show CoD can achieve a promising plasticity-stability trade-off and outperform existing diffusion-based methods and other representative baselines on most tasks.
comment: This work has been submitted to the IEEE for possible publication
♻ ☆ A Multi-Modal AI Copilot for Single-Cell Analysis with Instruction Following
Large language models excel at interpreting complex natural language instructions, enabling them to perform a wide range of tasks. In the life sciences, single-cell RNA sequencing (scRNA-seq) data serves as the "language of cellular biology", capturing intricate gene expression patterns at the single-cell level. However, interacting with this "language" through conventional tools is often inefficient and unintuitive, posing challenges for researchers. To address these limitations, we present InstructCell, a multi-modal AI copilot that leverages natural language as a medium for more direct and flexible single-cell analysis. We construct a comprehensive multi-modal instruction dataset that pairs text-based instructions with scRNA-seq profiles from diverse tissues and species. Building on this, we develop a multi-modal cell language architecture capable of simultaneously interpreting and processing both modalities. InstructCell empowers researchers to accomplish critical tasks-such as cell type annotation, conditional pseudo-cell generation, and drug sensitivity prediction-using straightforward natural language commands. Extensive evaluations demonstrate that InstructCell consistently meets or exceeds the performance of existing single-cell foundation models, while adapting to diverse experimental conditions. More importantly, InstructCell provides an accessible and intuitive tool for exploring complex single-cell data, lowering technical barriers and enabling deeper biological insights.
comment: 37 pages; 13 figures; Code: https://github.com/zjunlp/Instructcell, Models: https://huggingface.co/zjunlp/Instructcell-chat, https://huggingface.co/zjunlp/InstructCell-instruct
♻ ☆ Understanding Emergent Abilities of Language Models from the Loss Perspective NeurIPS 2024
Recent studies have put into question the belief that emergent abilities in language models are exclusive to large models. This skepticism arises from two observations: 1) smaller models can also exhibit high performance on emergent abilities and 2) there is doubt on the discontinuous metrics used to measure these abilities. In this paper, we propose to study emergent abilities in the lens of pre-training loss, instead of model size or training compute. We demonstrate that the Transformer models with the same pre-training loss, but different model and data sizes, generate the same performance on various downstream tasks, with a fixed data corpus, tokenization, and model architecture. We also discover that a model exhibits emergent abilities on certain tasks -- regardless of the continuity of metrics -- when its pre-training loss falls below a specific threshold. Before reaching this threshold, its performance remains at the level of random guessing. This inspires us to redefine emergent abilities as those that manifest in models with lower pre-training losses, highlighting that these abilities cannot be predicted by merely extrapolating the performance trends of models with higher pre-training losses.
comment: 23 pages, 8 figures. Accepted in NeurIPS 2024
♻ ☆ Data-driven inventory management for new products: A warm-start and adjusted Dyna-$Q$ approach
In this paper, we propose a novel reinforcement learning algorithm for inventory management of newly launched products with no or limited historical demand information. The algorithm follows the classic Dyna-$Q$ structure, balancing the model-based and model-free approaches, while accelerating the training process of Dyna-$Q$ and mitigating the model discrepancy generated by the model-based feedback. Warm-start information from the demand data of existing similar products can be incorporated into the algorithm to further stabilize the early-stage training and reduce the variance of the estimated optimal policy. Our approach is validated through a case study of bakery inventory management with real data. The adjusted Dyna-$Q$ shows up to a 23.7% reduction in average daily cost compared with $Q$-learning, and up to a 77.5% reduction in training time within the same horizon compared with classic Dyna-$Q$. By incorporating the warm-start information, it can be found that the adjusted Dyna-$Q$ has the lowest total cost, lowest variance in total cost, and relatively low shortage percentages among all the algorithms under a 30-day testing.
comment: 7 pages, 2 figures
♻ ☆ Unconditional stability of a recurrent neural circuit implementing divisive normalization
Stability in recurrent neural models poses a significant challenge, particularly in developing biologically plausible neurodynamical models that can be seamlessly trained. Traditional cortical circuit models are notoriously difficult to train due to expansive nonlinearities in the dynamical system, leading to an optimization problem with nonlinear stability constraints that are difficult to impose. Conversely, recurrent neural networks (RNNs) excel in tasks involving sequential data but lack biological plausibility and interpretability. In this work, we address these challenges by linking dynamic divisive normalization (DN) to the stability of ORGaNICs, a biologically plausible recurrent cortical circuit model that dynamically achieves DN and that has been shown to simulate a wide range of neurophysiological phenomena. By using the indirect method of Lyapunov, we prove the remarkable property of unconditional local stability for an arbitrary-dimensional ORGaNICs circuit when the recurrent weight matrix is the identity. We thus connect ORGaNICs to a system of coupled damped harmonic oscillators, which enables us to derive the circuit's energy function, providing a normative principle of what the circuit, and individual neurons, aim to accomplish. Further, for a generic recurrent weight matrix, we prove the stability of the 2D model and demonstrate empirically that stability holds in higher dimensions. Finally, we show that ORGaNICs can be trained by backpropagation through time without gradient clipping/scaling, thanks to its intrinsic stability property and adaptive time constants, which address the problems of exploding, vanishing, and oscillating gradients. By evaluating the model's performance on RNN benchmarks, we find that ORGaNICs outperform alternative neurodynamical models on static image classification tasks and perform comparably to LSTMs on sequential tasks.
♻ ☆ Investigating the Effect of Network Pruning on Performance and Interpretability
Deep Neural Networks (DNNs) are often over-parameterized for their tasks and can be compressed quite drastically by removing weights, a process called pruning. We investigate the impact of different pruning techniques on the classification performance and interpretability of GoogLeNet. We systematically apply unstructured and structured pruning, as well as connection sparsity (pruning of input weights) methods to the network and analyze the outcomes regarding the network's performance on the validation set of ImageNet. We also compare different retraining strategies, such as iterative pruning and one-shot pruning. We find that with sufficient retraining epochs, the performance of the networks can approximate the performance of the default GoogLeNet - and even surpass it in some cases. To assess interpretability, we employ the Mechanistic Interpretability Score (MIS) developed by Zimmermann et al. . Our experiments reveal that there is no significant relationship between interpretability and pruning rate when using MIS as a measure. Additionally, we observe that networks with extremely low accuracy can still achieve high MIS scores, suggesting that the MIS may not always align with intuitive notions of interpretability, such as understanding the basis of correct decisions.
comment: 4 pages, 6 figures
♻ ☆ Finite-Sample Bounds for Adaptive Inverse Reinforcement Learning using Passive Langevin Dynamics
This paper provides a finite-sample analysis of a passive stochastic gradient Langevin dynamics (PSGLD) algorithm. This algorithm is designed to achieve adaptive inverse reinforcement learning (IRL). Adaptive IRL aims to estimate the cost function of a forward learner performing a stochastic gradient algorithm (e.g., policy gradient reinforcement learning) by observing their estimates in real-time. The PSGLD algorithm is considered passive because it incorporates noisy gradients provided by an external stochastic gradient algorithm (forward learner), of which it has no control. The PSGLD algorithm acts as a randomized sampler to achieve adaptive IRL by reconstructing the forward learner's cost function nonparametrically from the stationary measure of a Langevin diffusion. This paper analyzes the non-asymptotic (finite-sample) performance; we provide explicit bounds on the 2-Wasserstein distance between PSGLD algorithm sample measure and the stationary measure encoding the cost function, and provide guarantees for a kernel density estimation scheme which reconstructs the cost function from empirical samples. Our analysis uses tools from the study of Markov diffusion operators. The derived bounds have both practical and theoretical significance. They provide finite-time guarantees for an adaptive IRL mechanism, and substantially generalize the analytical framework of a line of research in passive stochastic gradient algorithms.
♻ ☆ Compositional Automata Embeddings for Goal-Conditioned Reinforcement Learning
Goal-conditioned reinforcement learning is a powerful way to control an AI agent's behavior at runtime. That said, popular goal representations, e.g., target states or natural language, are either limited to Markovian tasks or rely on ambiguous task semantics. We propose representing temporal goals using compositions of deterministic finite automata (cDFAs) and use cDFAs to guide RL agents. cDFAs balance the need for formal temporal semantics with ease of interpretation: if one can understand a flow chart, one can understand a cDFA. On the other hand, cDFAs form a countably infinite concept class with Boolean semantics, and subtle changes to the automaton can result in very different tasks, making them difficult to condition agent behavior on. To address this, we observe that all paths through a DFA correspond to a series of reach-avoid tasks and propose pre-training graph neural network embeddings on "reach-avoid derived" DFAs. Through empirical evaluation, we demonstrate that the proposed pre-training method enables zero-shot generalization to various cDFA task classes and accelerated policy specialization without the myopic suboptimality of hierarchical methods.
♻ ☆ 2 OLMo 2 Furious
We present OLMo 2, the next generation of our fully open language models. OLMo 2 includes dense autoregressive models with improved architecture and training recipe, pretraining data mixtures, and instruction tuning recipes. Our modified model architecture and training recipe achieve both better training stability and improved per-token efficiency. Our updated pretraining data mixture introduces a new, specialized data mix called Dolmino Mix 1124, which significantly improves model capabilities across many downstream task benchmarks when introduced via late-stage curriculum training (i.e. specialized data during the annealing phase of pretraining). Finally, we incorporate best practices from T\"ulu 3 to develop OLMo 2-Instruct, focusing on permissive data and extending our final-stage reinforcement learning with verifiable rewards (RLVR). Our OLMo 2 base models sit at the Pareto frontier of performance to compute, often matching or outperforming open-weight only models like Llama 3.1 and Qwen 2.5 while using fewer FLOPs and with fully transparent training data, code, and recipe. Our fully open OLMo 2-Instruct models are competitive with or surpassing open-weight only models of comparable size, including Qwen 2.5, Llama 3.1 and Gemma 2. We release all OLMo 2 artifacts openly -- models at 7B and 13B scales, both pretrained and post-trained, including their full training data, training code and recipes, training logs and thousands of intermediate checkpoints. The final instruction model is available on the Ai2 Playground as a free research demo.
comment: Model demo available at playground.allenai.org
♻ ☆ Learning Cross-Domain Representations for Transferable Drug Perturbations on Single-Cell Transcriptional Responses AAAI
Phenotypic drug discovery has attracted widespread attention because of its potential to identify bioactive molecules. Transcriptomic profiling provides a comprehensive reflection of phenotypic changes in cellular responses to external perturbations. In this paper, we propose XTransferCDR, a novel generative framework designed for feature decoupling and transferable representation learning across domains. Given a pair of perturbed expression profiles, our approach decouples the perturbation representations from basal states through domain separation encoders and then cross-transfers them in the latent space. The transferred representations are then used to reconstruct the corresponding perturbed expression profiles via a shared decoder. This cross-transfer constraint effectively promotes the learning of transferable drug perturbation representations. We conducted extensive evaluations of our model on multiple datasets, including single-cell transcriptional responses to drugs and single- and combinatorial genetic perturbations. The experimental results show that XTransferCDR achieved better performance than current state-of-the-art methods, showcasing its potential to advance phenotypic drug discovery.
comment: Accepted by The 39th Annual AAAI Conference on Artificial Intelligenc (AAAI 2025)
♻ ☆ Efficient Long Video Tokenization via Coordinate-based Patch Reconstruction
Efficient tokenization of videos remains a challenge in training vision models that can process long videos. One promising direction is to develop a tokenizer that can encode long video clips, as it would enable the tokenizer to leverage the temporal coherence of videos better for tokenization. However, training existing tokenizers on long videos often incurs a huge training cost as they are trained to reconstruct all the frames at once. In this paper, we introduce CoordTok, a video tokenizer that learns a mapping from coordinate-based representations to the corresponding patches of input videos, inspired by recent advances in 3D generative models. In particular, CoordTok encodes a video into factorized triplane representations and reconstructs patches that correspond to randomly sampled $(x,y,t)$ coordinates. This allows for training large tokenizer models directly on long videos without requiring excessive training resources. Our experiments show that CoordTok can drastically reduce the number of tokens for encoding long video clips. For instance, CoordTok can encode a 128-frame video with 128$\times$128 resolution into 1280 tokens, while baselines need 6144 or 8192 tokens to achieve similar reconstruction quality. We further show that this efficient video tokenization enables memory-efficient training of a diffusion transformer that can generate 128 frames at once.
comment: Code is available on the project webpage: https://huiwon-jang.github.io/coordtok/
♻ ☆ A Unifying Information-theoretic Perspective on Evaluating Generative Models
Considering the difficulty of interpreting generative model output, there is significant current research focused on determining meaningful evaluation metrics. Several recent approaches utilize "precision" and "recall," borrowed from the classification domain, to individually quantify the output fidelity (realism) and output diversity (representation of the real data variation), respectively. With the increase in metric proposals, there is a need for a unifying perspective, allowing for easier comparison and clearer explanation of their benefits and drawbacks. To this end, we unify a class of kth-nearest-neighbors (kNN)-based metrics under an information-theoretic lens using approaches from kNN density estimation. Additionally, we propose a tri-dimensional metric composed of Precision Cross-Entropy (PCE), Recall Cross-Entropy (RCE), and Recall Entropy (RE), which separately measure fidelity and two distinct aspects of diversity, inter- and intra-class. Our domain-agnostic metric, derived from the information-theoretic concepts of entropy and cross-entropy, can be dissected for both sample- and mode-level analysis. Our detailed experimental results demonstrate the sensitivity of our metric components to their respective qualities and reveal undesirable behaviors of other metrics.
♻ ☆ Predicting Long-Term Student Outcomes from Short-Term EdTech Log Data
Educational stakeholders are often particularly interested in sparse, delayed student outcomes, like end-of-year statewide exams. The rare occurrence of such assessments makes it harder to identify students likely to fail such assessments, as well as making it slow for researchers and educators to be able to assess the effectiveness of particular educational tools. Prior work has primarily focused on using logs from students full usage (e.g. year-long) of an educational product to predict outcomes, or considered predictive accuracy using a few minutes to predict outcomes after a short (e.g. 1 hour) session. In contrast, we investigate machine learning predictors using students' logs during their first few hours of usage can provide useful predictive insight into those students' end-of-school year external assessment. We do this on three diverse datasets: from students in Uganda using a literacy game product, and from students in the US using two mathematics intelligent tutoring systems. We consider various measures of the accuracy of the resulting predictors, including its ability to identify students at different parts along the assessment performance distribution. Our findings suggest that short-term log usage data, from 2-5 hours, can be used to provide valuable signal about students' long-term external performance.
comment: Accepted to the 15th International Learning Analytics and Knowledge Conference (LAK2025)
♻ ☆ A Misclassification Network-Based Method for Comparative Genomic Analysis
Classifying genome sequences based on metadata has been an active area of research in comparative genomics for decades with many important applications across the life sciences. Established methods for classifying genomes can be broadly grouped into sequence alignment-based and alignment-free models. Conventional alignment-based models rely on genome similarity measures calculated based on local sequence alignments or consistent ordering among sequences. However, such methods are computationally expensive when dealing with large ensembles of even moderately sized genomes. In contrast, alignment-free (AF) approaches measure genome similarity based on summary statistics in an unsupervised setting and are efficient enough to analyze large datasets. However, both alignment-based and AF methods typically assume fixed scoring rubrics that lack the flexibility to assign varying importance to different parts of the sequences based on prior knowledge. In this study, we integrate AI and network science approaches to develop a comparative genomic analysis framework that addresses these limitations. Our approach, termed the Genome Misclassification Network Analysis (GMNA), simultaneously leverages misclassified instances, a learned scoring rubric, and label information to classify genomes based on associated metadata and better understand potential drivers of misclassification. We evaluate the utility of the GMNA using Naive Bayes and convolutional neural network models, supplemented by additional experiments with transformer-based models, to construct SARS-CoV-2 sampling location classifiers using over 500,000 viral genome sequences and study the resulting network of misclassifications. We demonstrate the global health potential of the GMNA by leveraging the SARS-CoV-2 genome misclassification networks to investigate the role human mobility played in structuring geographic clustering of SARS-CoV-2.
♻ ☆ FlowDock: Geometric Flow Matching for Generative Protein-Ligand Docking and Affinity Prediction
Powerful generative AI models of protein-ligand structure have recently been proposed, but few of these methods support both flexible protein-ligand docking and affinity estimation. Of those that do, none can directly model multiple binding ligands concurrently or have been rigorously benchmarked on pharmacologically relevant drug targets, hindering their widespread adoption in drug discovery efforts. In this work, we propose FlowDock, the first deep geometric generative model based on conditional flow matching that learns to directly map unbound (apo) structures to their bound (holo) counterparts for an arbitrary number of binding ligands. Furthermore, FlowDock provides predicted structural confidence scores and binding affinity values with each of its generated protein-ligand complex structures, enabling fast virtual screening of new (multi-ligand) drug targets. For the well-known PoseBusters Benchmark dataset, FlowDock outperforms single-sequence AlphaFold 3 with a 51% blind docking success rate using unbound (apo) protein input structures and without any information derived from multiple sequence alignments, and for the challenging new DockGen-E dataset, FlowDock outperforms single-sequence AlphaFold 3 and matches single-sequence Chai-1 for binding pocket generalization. Additionally, in the ligand category of the 16th community-wide Critical Assessment of Techniques for Structure Prediction (CASP16), FlowDock ranked among the top-5 methods for pharmacological binding affinity estimation across 140 protein-ligand complexes, demonstrating the efficacy of its learned representations in virtual screening. Source code, data, and pre-trained models are available at https://github.com/BioinfoMachineLearning/FlowDock.
comment: 10 pages, 2 tables, 2 algorithms, 7 figures. Code, data, pre-trained models, and baseline method predictions are available at https://github.com/BioinfoMachineLearning/FlowDock
♻ ☆ Top-k Multi-Armed Bandit Learning for Content Dissemination in Swarms of Micro-UAVs
This paper presents a Micro-Unmanned Aerial Vehicle (UAV)-enhanced content management system for disaster scenarios where communication infrastructure is generally compromised. Utilizing a hybrid network of stationary and mobile Micro-UAVs, this system aims to provide crucial content access to isolated communities. In the developed architecture, stationary anchor UAVs, equipped with vertical and lateral links, serve users in individual disaster-affected communities. and mobile micro-ferrying UAVs, with enhanced mobility, extend coverage across multiple such communities. The primary goal is to devise a content dissemination system that dynamically learns caching policies to maximize content accessibility to users left without communication infrastructure. The core contribution is an adaptive content dissemination framework that employs a decentralized Top-k Multi-Armed Bandit learning approach for efficient UAV caching decisions. This approach accounts for geo-temporal variations in content popularity and diverse user demands. Additionally, a Selective Caching Algorithm is proposed to minimize redundant content copies by leveraging inter-UAV information sharing. Through functional verification and performance evaluation, the proposed framework demonstrates improved system performance and adaptability across varying network sizes, micro-UAV swarms, and content popularity distributions.
comment: 16 pages, 8 figures, 2 algorithms, 2 tables, journal
♻ ☆ Key-Exchange Convolutional Auto-Encoder for Data Augmentation in Early Knee Osteoarthritis Detection
Knee Osteoarthritis (KOA) is a common musculoskeletal condition that significantly affects mobility and quality of life, particularly in elderly populations. However, training deep learning models for early KOA classification is often hampered by the limited availability of annotated medical datasets, owing to the high costs and labour-intensive nature of data labelling. Traditional data augmentation techniques, while useful, rely on simple transformations and fail to introduce sufficient diversity into the dataset. To address these challenges, we propose the Key-Exchange Convolutional Auto-Encoder (KECAE) as an innovative Artificial Intelligence (AI)-based data augmentation strategy for early KOA classification. Our model employs a convolutional autoencoder with a novel key-exchange mechanism that generates synthetic images by selectively exchanging key pathological features between X-ray images, which not only diversifies the dataset but also ensures the clinical validity of the augmented data. A hybrid loss function is introduced to supervise feature learning and reconstruction, integrating multiple components, including reconstruction, supervision, and feature separation losses. Experimental results demonstrate that the KECAE-generated data significantly improve the performance of KOA classification models, with accuracy gains of up to 1.98% across various standard and state-of-the-art architectures. Furthermore, a clinical validation study involving expert radiologists confirms the anatomical plausibility and diagnostic realism of the synthetic outputs. These findings highlight the potential of KECAE as a robust tool for augmenting medical datasets in early KOA detection.
♻ ☆ Nonsmooth Nonconvex-Nonconcave Minimax Optimization: Primal-Dual Balancing and Iteration Complexity Analysis
Nonconvex-nonconcave minimax optimization has gained widespread interest over the last decade. However, most existing works focus on variants of gradient descent-ascent (GDA) algorithms, which are only applicable to smooth nonconvex-concave settings. To address this limitation, we propose a novel algorithm named smoothed proximal linear descent-ascent (smoothed PLDA), which can effectively handle a broad range of structured nonsmooth nonconvex-nonconcave minimax problems. Specifically, we consider the setting where the primal function has a nonsmooth composite structure and the dual function possesses the Kurdyka-Lojasiewicz (KL) property with exponent $\theta \in [0,1)$. We introduce a novel convergence analysis framework for smoothed PLDA, the key components of which are our newly developed nonsmooth primal error bound and dual error bound. Using this framework, we show that smoothed PLDA can find both $\epsilon$-game-stationary points and $\epsilon$-optimization-stationary points of the problems of interest in $\mathcal{O}(\epsilon^{-2\max\{2\theta,1\}})$ iterations. Furthermore, when $\theta \in [0,\frac{1}{2}]$, smoothed PLDA achieves the optimal iteration complexity of $\mathcal{O}(\epsilon^{-2})$. To further demonstrate the effectiveness and wide applicability of our analysis framework, we show that certain max-structured problem possesses the KL property with exponent $\theta=0$ under mild assumptions. As a by-product, we establish algorithm-independent quantitative relationships among various stationarity concepts, which may be of independent interest.
comment: Accepted for publication in Mathematical Programming
Artificial Intelligence 137
☆ How Do Generative Models Draw a Software Engineer? A Case Study on Stable Diffusion Bias
Generative models are nowadays widely used to generate graphical content used for multiple purposes, e.g. web, art, advertisement. However, it has been shown that the images generated by these models could reinforce societal biases already existing in specific contexts. In this paper, we focus on understanding if this is the case when one generates images related to various software engineering tasks. In fact, the Software Engineering (SE) community is not immune from gender and ethnicity disparities, which could be amplified by the use of these models. Hence, if used without consciousness, artificially generated images could reinforce these biases in the SE domain. Specifically, we perform an extensive empirical evaluation of the gender and ethnicity bias exposed by three versions of the Stable Diffusion (SD) model (a very popular open-source text-to-image model) - SD 2, SD XL, and SD 3 - towards SE tasks. We obtain 6,720 images by feeding each model with two sets of prompts describing different software-related tasks: one set includes the Software Engineer keyword, and one set does not include any specification of the person performing the task. Next, we evaluate the gender and ethnicity disparities in the generated images. Results show how all models are significantly biased towards male figures when representing software engineers. On the contrary, while SD 2 and SD XL are strongly biased towards White figures, SD 3 is slightly more biased towards Asian figures. Nevertheless, all models significantly under-represent Black and Arab figures, regardless of the prompt style used. The results of our analysis highlight severe concerns about adopting those models to generate content for SE tasks and open the field for future research on bias mitigation in this context.
☆ Multimodal LLMs Can Reason about Aesthetics in Zero-Shot
We present the first study on how Multimodal LLMs' (MLLMs) reasoning ability shall be elicited to evaluate the aesthetics of artworks. To facilitate this investigation, we construct MM-StyleBench, a novel high-quality dataset for benchmarking artistic stylization. We then develop a principled method for human preference modeling and perform a systematic correlation analysis between MLLMs' responses and human preference. Our experiments reveal an inherent hallucination issue of MLLMs in art evaluation, associated with response subjectivity. ArtCoT is proposed, demonstrating that art-specific task decomposition and the use of concrete language boost MLLMs' reasoning ability for aesthetics. Our findings offer valuable insights into MLLMs for art and can benefit a wide range of downstream applications, such as style transfer and artistic image generation. Code available at https://github.com/songrise/MLLM4Art.
comment: WIP, Homepage https://github.com/songrise/MLLM4Art
☆ AI-RAN: Transforming RAN with AI-driven Computing Infrastructure
The radio access network (RAN) landscape is undergoing a transformative shift from traditional, communication-centric infrastructures towards converged compute-communication platforms. This article introduces AI-RAN which integrates both RAN and artificial intelligence (AI) workloads on the same infrastructure. By doing so, AI-RAN not only meets the performance demands of future networks but also improves asset utilization. We begin by examining how RANs have evolved beyond mobile broadband towards AI-RAN and articulating manifestations of AI-RAN into three forms: AI-for-RAN, AI-on-RAN, and AI-and-RAN. Next, we identify the key requirements and enablers for the convergence of communication and computing in AI-RAN. We then provide a reference architecture for advancing AI-RAN from concept to practice. To illustrate the practical potential of AI-RAN, we present a proof-of-concept that concurrently processes RAN and AI workloads utilizing NVIDIA Grace-Hopper GH200 servers. Finally, we conclude the article by outlining future work directions to guide further developments of AI-RAN.
comment: 7 pages, 5 figures
☆ Personality Modeling for Persuasion of Misinformation using AI Agent
The proliferation of misinformation on social media platforms has highlighted the need to understand how individual personality traits influence susceptibility to and propagation of misinformation. This study employs an innovative agent-based modeling approach to investigate the relationship between personality traits and misinformation dynamics. Using six AI agents embodying different dimensions of the Big Five personality traits (Extraversion, Agreeableness, and Neuroticism), we simulated interactions across six diverse misinformation topics. The experiment, implemented through the AgentScope framework using the GLM-4-Flash model, generated 90 unique interactions, revealing complex patterns in how personality combinations affect persuasion and resistance to misinformation. Our findings demonstrate that analytical and critical personality traits enhance effectiveness in evidence-based discussions, while non-aggressive persuasion strategies show unexpected success in misinformation correction. Notably, agents with critical traits achieved a 59.4% success rate in HIV-related misinformation discussions, while those employing non-aggressive approaches maintained consistent persuasion rates above 40% across different personality combinations. The study also revealed a non-transitive pattern in persuasion effectiveness, challenging conventional assumptions about personality-based influence. These results provide crucial insights for developing personality-aware interventions in digital environments and suggest that effective misinformation countermeasures should prioritize emotional connection and trust-building over confrontational approaches. The findings contribute to both theoretical understanding of personality-misinformation dynamics and practical strategies for combating misinformation in social media contexts.
☆ Development and Validation of the Provider Documentation Summarization Quality Instrument for Large Language Models
As Large Language Models (LLMs) are integrated into electronic health record (EHR) workflows, validated instruments are essential to evaluate their performance before implementation. Existing instruments for provider documentation quality are often unsuitable for the complexities of LLM-generated text and lack validation on real-world data. The Provider Documentation Summarization Quality Instrument (PDSQI-9) was developed to evaluate LLM-generated clinical summaries. Multi-document summaries were generated from real-world EHR data across multiple specialties using several LLMs (GPT-4o, Mixtral 8x7b, and Llama 3-8b). Validation included Pearson correlation for substantive validity, factor analysis and Cronbach's alpha for structural validity, inter-rater reliability (ICC and Krippendorff's alpha) for generalizability, a semi-Delphi process for content validity, and comparisons of high- versus low-quality summaries for discriminant validity. Seven physician raters evaluated 779 summaries and answered 8,329 questions, achieving over 80% power for inter-rater reliability. The PDSQI-9 demonstrated strong internal consistency (Cronbach's alpha = 0.879; 95% CI: 0.867-0.891) and high inter-rater reliability (ICC = 0.867; 95% CI: 0.867-0.868), supporting structural validity and generalizability. Factor analysis identified a 4-factor model explaining 58% of the variance, representing organization, clarity, accuracy, and utility. Substantive validity was supported by correlations between note length and scores for Succinct (rho = -0.200, p = 0.029) and Organized (rho = -0.190, p = 0.037). Discriminant validity distinguished high- from low-quality summaries (p < 0.001). The PDSQI-9 demonstrates robust construct validity, supporting its use in clinical practice to evaluate LLM-generated summaries and facilitate safer integration of LLMs into healthcare workflows.
☆ Trusted Machine Learning Models Unlock Private Inference for Problems Currently Infeasible with Cryptography
We often interact with untrusted parties. Prioritization of privacy can limit the effectiveness of these interactions, as achieving certain goals necessitates sharing private data. Traditionally, addressing this challenge has involved either seeking trusted intermediaries or constructing cryptographic protocols that restrict how much data is revealed, such as multi-party computations or zero-knowledge proofs. While significant advances have been made in scaling cryptographic approaches, they remain limited in terms of the size and complexity of applications they can be used for. In this paper, we argue that capable machine learning models can fulfill the role of a trusted third party, thus enabling secure computations for applications that were previously infeasible. In particular, we describe Trusted Capable Model Environments (TCMEs) as an alternative approach for scaling secure computation, where capable machine learning model(s) interact under input/output constraints, with explicit information flow control and explicit statelessness. This approach aims to achieve a balance between privacy and computational efficiency, enabling private inference where classical cryptographic solutions are currently infeasible. We describe a number of use cases that are enabled by TCME, and show that even some simple classic cryptographic problems can already be solved with TCME. Finally, we outline current limitations and discuss the path forward in implementing them.
☆ An analysis of data variation and bias in image-based dermatological datasets for machine learning classification
AI algorithms have become valuable in aiding professionals in healthcare. The increasing confidence obtained by these models is helpful in critical decision demands. In clinical dermatology, classification models can detect malignant lesions on patients' skin using only RGB images as input. However, most learning-based methods employ data acquired from dermoscopic datasets on training, which are large and validated by a gold standard. Clinical models aim to deal with classification on users' smartphone cameras that do not contain the corresponding resolution provided by dermoscopy. Also, clinical applications bring new challenges. It can contain captures from uncontrolled environments, skin tone variations, viewpoint changes, noises in data and labels, and unbalanced classes. A possible alternative would be to use transfer learning to deal with the clinical images. However, as the number of samples is low, it can cause degradations on the model's performance; the source distribution used in training differs from the test set. This work aims to evaluate the gap between dermoscopic and clinical samples and understand how the dataset variations impact training. It assesses the main differences between distributions that disturb the model's prediction. Finally, from experiments on different architectures, we argue how to combine the data from divergent distributions, decreasing the impact on the model's final accuracy.
comment: 10 pages, 1 figure
☆ Kolmogorov-Arnold Networks for Time Series Granger Causality Inference
We introduce Granger Causality Kolmogorov-Arnold Networks (GCKAN), an innovative architecture that extends the recently proposed Kolmogorov-Arnold Networks (KAN) to the domain of causal inference. By extracting base weights from KAN layers and incorporating the sparsity-inducing penalty along with ridge regularization, GCKAN infers the Granger causality from time series while enabling automatic time lag selection. Additionally, we propose an algorithm leveraging time-reversed Granger causality to enhance inference accuracy. The algorithm compares prediction and sparse-inducing losses derived from the original and time-reversed series, automatically selecting the casual relationship with the higher score or integrating the results to mitigate spurious connectivities. Comprehensive experiments conducted on Lorenz-96, gene regulatory networks, fMRI BOLD signals, and VAR datasets demonstrate that the proposed model achieves competitive performance to state-of-the-art methods in inferring Granger causality from nonlinear, high-dimensional, and limited-sample time series.
☆ Analyzing the Ethical Logic of Six Large Language Models
This study examines the ethical reasoning of six prominent generative large language models: OpenAI GPT-4o, Meta LLaMA 3.1, Perplexity, Anthropic Claude 3.5 Sonnet, Google Gemini, and Mistral 7B. The research explores how these models articulate and apply ethical logic, particularly in response to moral dilemmas such as the Trolley Problem, and Heinz Dilemma. Departing from traditional alignment studies, the study adopts an explainability-transparency framework, prompting models to explain their ethical reasoning. This approach is analyzed through three established ethical typologies: the consequentialist-deontological analytic, Moral Foundations Theory, and the Kohlberg Stages of Moral Development Model. Findings reveal that LLMs exhibit largely convergent ethical logic, marked by a rationalist, consequentialist emphasis, with decisions often prioritizing harm minimization and fairness. Despite similarities in pre-training and model architecture, a mixture of nuanced and significant differences in ethical reasoning emerge across models, reflecting variations in fine-tuning and post-training processes. The models consistently display erudition, caution, and self-awareness, presenting ethical reasoning akin to a graduate-level discourse in moral philosophy. In striking uniformity these systems all describe their ethical reasoning as more sophisticated than what is characteristic of typical human moral logic.
☆ Visual WetlandBirds Dataset: Bird Species Identification and Behavior Recognition in Videos
The current biodiversity loss crisis makes animal monitoring a relevant field of study. In light of this, data collected through monitoring can provide essential insights, and information for decision-making aimed at preserving global biodiversity. Despite the importance of such data, there is a notable scarcity of datasets featuring videos of birds, and none of the existing datasets offer detailed annotations of bird behaviors in video format. In response to this gap, our study introduces the first fine-grained video dataset specifically designed for bird behavior detection and species classification. This dataset addresses the need for comprehensive bird video datasets and provides detailed data on bird actions, facilitating the development of deep learning models to recognize these, similar to the advancements made in human action recognition. The proposed dataset comprises 178 videos recorded in Spanish wetlands, capturing 13 different bird species performing 7 distinct behavior classes. In addition, we also present baseline results using state of the art models on two tasks: bird behavior recognition and species classification.
☆ Disentangling Exploration of Large Language Models by Optimal Exploitation
Exploration is a crucial skill for self-improvement and open-ended problem-solving. However, it remains uncertain whether large language models can effectively explore the state-space. Existing evaluations predominantly focus on the trade-off between exploration and exploitation, often assessed in multi-armed bandit problems. In contrast, this work isolates exploration as the sole objective, tasking the agent with delivering information that enhances future returns. For the evaluation, we propose to decompose missing rewards into exploration and exploitation components by measuring the optimal achievable return for the states already explored. Our experiments with various LLMs reveal that most models struggle to sufficiently explore the state-space and that weak exploration is insufficient. We observe a positive correlation between model size and exploration performance, with larger models demonstrating superior capabilities. Furthermore, we show that our decomposition provides insights into differences in behaviors driven by agent instructions during prompt engineering, offering a valuable tool for refining LLM performance in exploratory tasks.
☆ Modeling Melt Pool Features and Spatter Using Symbolic Regression and Machine Learning
Additive manufacturing (AM) is a rapidly evolving technology that has attracted applications across a wide range of fields due to its ability to fabricate complex geometries. However, one of the key challenges in AM is achieving consistent print quality. This inconsistency is often attributed to uncontrolled melt pool dynamics, partly caused by spatter which can lead to defects. Therefore, capturing and controlling the evolution of the melt pool is crucial for enhancing process stability and part quality. In this study, we developed a framework to support decision-making in AM operations, facilitating quality control and minimizing defects via machine learning (ML) and polynomial symbolic regression models. We implemented experimentally validated computational tools as a cost-effective approach to collect large datasets from laser powder bed fusion (LPBF) processes. For a dataset consisting of 281 process conditions, parameters such as melt pool dimensions (length, width, depth), melt pool geometry (area, volume), and volume indicated as spatter were extracted. Using machine learning (ML) and polynomial symbolic regression models, a high R2 of over 95 % was achieved in predicting the melt pool dimensions and geometry features for both the training and testing datasets, with either process conditions (power and velocity) or melt pool dimensions as the model inputs. In the case of volume indicated as spatter, R2 improved after logarithmic transforming the model inputs, which was either the process conditions or the melt pool dimensions. Among the investigated ML models, the ExtraTree model achieved the highest R2 values of 96.7 % and 87.5 %.
☆ Projection Implicit Q-Learning with Support Constraint for Offline Reinforcement Learning
Offline Reinforcement Learning (RL) faces a critical challenge of extrapolation errors caused by out-of-distribution (OOD) actions. Implicit Q-Learning (IQL) algorithm employs expectile regression to achieve in-sample learning, effectively mitigating the risks associated with OOD actions. However, the fixed hyperparameter in policy evaluation and density-based policy improvement method limit its overall efficiency. In this paper, we propose Proj-IQL, a projective IQL algorithm enhanced with the support constraint. In the policy evaluation phase, Proj-IQL generalizes the one-step approach to a multi-step approach through vector projection, while maintaining in-sample learning and expectile regression framework. In the policy improvement phase, Proj-IQL introduces support constraint that is more aligned with the policy evaluation approach. Furthermore, we theoretically demonstrate that Proj-IQL guarantees monotonic policy improvement and enjoys a progressively more rigorous criterion for superior actions. Empirical results demonstrate the Proj-IQL achieves state-of-the-art performance on D4RL benchmarks, especially in challenging navigation domains.
☆ Computing Game Symmetries and Equilibria That Respect Them AAAI
Strategic interactions can be represented more concisely, and analyzed and solved more efficiently, if we are aware of the symmetries within the multiagent system. Symmetries also have conceptual implications, for example for equilibrium selection. We study the computational complexity of identifying and using symmetries. Using the classical framework of normal-form games, we consider game symmetries that can be across some or all players and/or actions. We find a strong connection between game symmetries and graph automorphisms, yielding graph automorphism and graph isomorphism completeness results for characterizing the symmetries present in a game. On the other hand, we also show that the problem becomes polynomial-time solvable when we restrict the consideration of actions in one of two ways. Next, we investigate when exactly game symmetries can be successfully leveraged for Nash equilibrium computation. We show that finding a Nash equilibrium that respects a given set of symmetries is PPAD- and CLS-complete in general-sum and team games respectively -- that is, exactly as hard as Brouwer fixed point and gradient descent problems. Finally, we present polynomial-time methods for the special cases where we are aware of a vast number of symmetries, or where the game is two-player zero-sum and we do not even know the symmetries.
comment: Long and updated version to the published paper in the Proceedings of the 39th Annual AAAI Conference on Artificial Intelligence (AAAI 2025). 24 pages, 2 figures, 1 table
☆ Leveraging Large Language Models as Knowledge-Driven Agents for Reliable Retrosynthesis Planning
Identifying reliable synthesis pathways in materials chemistry is a complex task, particularly in polymer science, due to the intricate and often non-unique nomenclature of macromolecules. To address this challenge, we propose an agent system that integrates large language models (LLMs) and knowledge graphs (KGs). By leveraging LLMs' powerful capabilities for extracting and recognizing chemical substance names, and storing the extracted data in a structured knowledge graph, our system fully automates the retrieval of relevant literatures, extraction of reaction data, database querying, construction of retrosynthetic pathway trees, further expansion through the retrieval of additional literature and recommendation of optimal reaction pathways. A novel Multi-branched Reaction Pathway Search (MBRPS) algorithm enables the exploration of all pathways, with a particular focus on multi-branched ones, helping LLMs overcome weak reasoning in multi-branched paths. This work represents the first attempt to develop a fully automated retrosynthesis planning agent tailored specially for macromolecules powered by LLMs. Applied to polyimide synthesis, our new approach constructs a retrosynthetic pathway tree with hundreds of pathways and recommends optimized routes, including both known and novel pathways, demonstrating its effectiveness and potential for broader applications.
☆ Karatsuba Matrix Multiplication and its Efficient Custom Hardware Implementations
While the Karatsuba algorithm reduces the complexity of large integer multiplication, the extra additions required minimize its benefits for smaller integers of more commonly-used bitwidths. In this work, we propose the extension of the scalar Karatsuba multiplication algorithm to matrix multiplication, showing how this maintains the reduction in multiplication complexity of the original Karatsuba algorithm while reducing the complexity of the extra additions. Furthermore, we propose new matrix multiplication hardware architectures for efficiently exploiting this extension of the Karatsuba algorithm in custom hardware. We show that the proposed algorithm and hardware architectures can provide real area or execution time improvements for integer matrix multiplication compared to scalar Karatsuba or conventional matrix multiplication algorithms, while also supporting implementation through proven systolic array and conventional multiplier architectures at the core. We provide a complexity analysis of the algorithm and architectures and evaluate the proposed designs both in isolation and in an end-to-end deep learning accelerator system compared to baseline designs and prior state-of-the-art works implemented on the same type of compute platform, demonstrating their ability to increase the performance-per-area of matrix multiplication hardware.
comment: Accepted for publication in IEEE Transactions on Computers; Associated source code available on github at https://github.com/trevorpogue/algebraic-nnhw
☆ Incrementally Learning Multiple Diverse Data Domains via Multi-Source Dynamic Expansion Model
Continual Learning seeks to develop a model capable of incrementally assimilating new information while retaining prior knowledge. However, current research predominantly addresses a straightforward learning context, wherein all data samples originate from a singular data domain. This paper shifts focus to a more complex and realistic learning environment, characterized by data samples sourced from multiple distinct domains. We tackle this intricate learning challenge by introducing a novel methodology, termed the Multi-Source Dynamic Expansion Model (MSDEM), which leverages various pre-trained models as backbones and progressively establishes new experts based on them to adapt to emerging tasks. Additionally, we propose an innovative dynamic expandable attention mechanism designed to selectively harness knowledge from multiple backbones, thereby accelerating the new task learning. Moreover, we introduce a dynamic graph weight router that strategically reuses all previously acquired parameters and representations for new task learning, maximizing the positive knowledge transfer effect, which further improves generalization performance. We conduct a comprehensive series of experiments, and the empirical findings indicate that our proposed approach achieves state-of-the-art performance.
comment: 10 pages, 5 figures
☆ Silent Abandonment in Text-Based Contact Centers: Identifying, Quantifying, and Mitigating its Operational Impacts
In the quest to improve services, companies offer customers the option to interact with agents via texting. Such contact centers face unique challenges compared to traditional call centers, as measuring customer experience proxies like abandonment and patience involves uncertainty. A key source of this uncertainty is silent abandonment, where customers leave without notifying the system, wasting agent time and leaving their status unclear. Silent abandonment also obscures whether a customer was served or left. Our goals are to measure the magnitude of silent abandonment and mitigate its effects. Classification models show that 3%-70% of customers across 17 companies abandon silently. In one study, 71.3% of abandoning customers did so silently, reducing agent efficiency by 3.2% and system capacity by 15.3%, incurring $5,457 in annual costs per agent. We develop an expectation-maximization (EM) algorithm to estimate customer patience under uncertainty and identify influencing covariates. We find that companies should use classification models to estimate abandonment scope and our EM algorithm to assess patience. We suggest strategies to operationally mitigate the impact of silent abandonment by predicting suspected silent-abandonment behavior or changing service design. Specifically, we show that while allowing customers to write while waiting in the queue creates a missing data challenge, it also significantly increases patience and reduces service time, leading to reduced abandonment and lower staffing requirements.
comment: arXiv admin note: text overlap with arXiv:2304.11754
☆ ARMOR: Shielding Unlearnable Examples against Data Augmentation
Private data, when published online, may be collected by unauthorized parties to train deep neural networks (DNNs). To protect privacy, defensive noises can be added to original samples to degrade their learnability by DNNs. Recently, unlearnable examples are proposed to minimize the training loss such that the model learns almost nothing. However, raw data are often pre-processed before being used for training, which may restore the private information of protected data. In this paper, we reveal the data privacy violation induced by data augmentation, a commonly used data pre-processing technique to improve model generalization capability, which is the first of its kind as far as we are concerned. We demonstrate that data augmentation can significantly raise the accuracy of the model trained on unlearnable examples from 21.3% to 66.1%. To address this issue, we propose a defense framework, dubbed ARMOR, to protect data privacy from potential breaches of data augmentation. To overcome the difficulty of having no access to the model training process, we design a non-local module-assisted surrogate model that better captures the effect of data augmentation. In addition, we design a surrogate augmentation selection strategy that maximizes distribution alignment between augmented and non-augmented samples, to choose the optimal augmentation strategy for each class. We also use a dynamic step size adjustment algorithm to enhance the defensive noise generation process. Extensive experiments are conducted on 4 datasets and 5 data augmentation methods to verify the performance of ARMOR. Comparisons with 6 state-of-the-art defense methods have demonstrated that ARMOR can preserve the unlearnability of protected private data under data augmentation. ARMOR reduces the test accuracy of the model trained on augmented protected samples by as much as 60% more than baselines.
☆ Digital Phenotyping for Adolescent Mental Health: A Feasibility Study Employing Machine Learning to Predict Mental Health Risk From Active and Passive Smartphone Data
Background: Adolescents are particularly vulnerable to mental disorders, with over 75% of cases manifesting before the age of 25. Research indicates that only 18 to 34% of young people experiencing high levels of depression or anxiety symptoms seek support. Digital tools leveraging smartphones offer scalable and early intervention opportunities. Objective: Using a novel machine learning framework, this study evaluated the feasibility of integrating active and passive smartphone data to predict mental disorders in non-clinical adolescents. Specifically, we investigated the utility of the Mindcraft app in predicting risks for internalising and externalising disorders, eating disorders, insomnia and suicidal ideation. Methods: Participants (N=103; mean age 16.1 years) were recruited from three London schools. Participants completed the Strengths and Difficulties Questionnaire, the Eating Disorders-15 Questionnaire, Sleep Condition Indicator Questionnaire and indicated the presence/absence of suicidal ideation. They used the Mindcraft app for 14 days, contributing active data via self-reports and passive data from smartphone sensors. A contrastive pretraining phase was applied to enhance user-specific feature stability, followed by supervised fine-tuning. The model evaluation employed leave-one-subject-out cross-validation using balanced accuracy as the primary metric. Results: The integration of active and passive data achieved superior performance compared to individual data sources, with mean balanced accuracies of 0.71 for SDQ-High risk, 0.67 for insomnia, 0.77 for suicidal ideation and 0.70 for eating disorders. The contrastive learning framework stabilised daily behavioural representations, enhancing predictive robustness. This study demonstrates the potential of integrating active and passive smartphone data with advanced machine-learning techniques for predicting mental health risks.
☆ Graph Counterfactual Explainable AI via Latent Space Traversal
Explaining the predictions of a deep neural network is a nontrivial task, yet high-quality explanations for predictions are often a prerequisite for practitioners to trust these models. Counterfactual explanations aim to explain predictions by finding the ''nearest'' in-distribution alternative input whose prediction changes in a pre-specified way. However, it remains an open question how to define this nearest alternative input, whose solution depends on both the domain (e.g. images, graphs, tabular data, etc.) and the specific application considered. For graphs, this problem is complicated i) by their discrete nature, as opposed to the continuous nature of state-of-the-art graph classifiers; and ii) by the node permutation group acting on the graphs. We propose a method to generate counterfactual explanations for any differentiable black-box graph classifier, utilizing a case-specific permutation equivariant graph variational autoencoder. We generate counterfactual explanations in a continuous fashion by traversing the latent space of the autoencoder across the classification boundary of the classifier, allowing for seamless integration of discrete graph structure and continuous graph attributes. We empirically validate the approach on three graph datasets, showing that our model is consistently high-performing and more robust than the baselines.
comment: Published at Northern Lights Deep Learning Conference 2025
☆ RouteNet-Gauss: Hardware-Enhanced Network Modeling with Machine Learning
Network simulation is pivotal in network modeling, assisting with tasks ranging from capacity planning to performance estimation. Traditional approaches such as Discrete Event Simulation (DES) face limitations in terms of computational cost and accuracy. This paper introduces RouteNet-Gauss, a novel integration of a testbed network with a Machine Learning (ML) model to address these challenges. By using the testbed as a hardware accelerator, RouteNet-Gauss generates training datasets rapidly and simulates network scenarios with high fidelity to real-world conditions. Experimental results show that RouteNet-Gauss significantly reduces prediction errors by up to 95% and achieves a 488x speedup in inference time compared to state-of-the-art DES-based methods. RouteNet-Gauss's modular architecture is dynamically constructed based on the specific characteristics of the network scenario, such as topology and routing. This enables it to understand and generalize to different network configurations beyond those seen during training, including networks up to 10x larger. Additionally, it supports Temporal Aggregated Performance Estimation (TAPE), providing configurable temporal granularity and maintaining high accuracy in flow performance metrics. This approach shows promise in improving both simulation efficiency and accuracy, offering a valuable tool for network operators.
comment: 13 pages, 11 figures
☆ Automatic tuning of communication protocols for vehicular ad hoc networks using metaheuristics
The emerging field of vehicular ad hoc networks (VANETs) deals with a set of communicating vehicles which are able to spontaneously interconnect without any pre-existing infrastructure. In such kind of networks, it is crucial to make an optimal configuration of the communication protocols previously to the final network deployment. This way, a human designer can obtain an optimal QoS of the network beforehand. The problem we consider in this work lies in configuring the File Transfer protocol Configuration (FTC) with the aim of optimizing the transmission time, the number of lost packets, and the amount of data transferred in realistic VANET scenarios. We face the FTC with five representative state-of-the-art optimization techniques and compare their performance. These algorithms are: Particle Swarm Optimization (PSO), Differential Evolution (DE), Genetic Algorithm (GA), Evolutionary Strategy (ES), and Simulated Annealing (SA). For our tests, two typical environment instances of VANETs for Urban and Highway scenarios have been defined. The experiments using ns- 2 (a well-known realistic VANET simulator) reveal that PSO outperforms all the compared algorithms for both studied VANET instances.
☆ Exploring Task-Level Optimal Prompts for Visual In-Context Learning
With the development of Vision Foundation Models (VFMs) in recent years, Visual In-Context Learning (VICL) has become a better choice compared to modifying models in most scenarios. Different from retraining or fine-tuning model, VICL does not require modifications to the model's weights or architecture, and only needs a prompt with demonstrations to teach VFM how to solve tasks. Currently, significant computational cost for finding optimal prompts for every test sample hinders the deployment of VICL, as determining which demonstrations to use for constructing prompts is very costly. In this paper, however, we find a counterintuitive phenomenon that most test samples actually achieve optimal performance under the same prompts, and searching for sample-level prompts only costs more time but results in completely identical prompts. Therefore, we propose task-level prompting to reduce the cost of searching for prompts during the inference stage and introduce two time-saving yet effective task-level prompt search strategies. Extensive experimental results show that our proposed method can identify near-optimal prompts and reach the best VICL performance with a minimal cost that prior work has never achieved.
☆ ToMATO: Verbalizing the Mental States of Role-Playing LLMs for Benchmarking Theory of Mind AAAI 2025
Existing Theory of Mind (ToM) benchmarks diverge from real-world scenarios in three aspects: 1) they assess a limited range of mental states such as beliefs, 2) false beliefs are not comprehensively explored, and 3) the diverse personality traits of characters are overlooked. To address these challenges, we introduce ToMATO, a new ToM benchmark formulated as multiple-choice QA over conversations. ToMATO is generated via LLM-LLM conversations featuring information asymmetry. By employing a prompting method that requires role-playing LLMs to verbalize their thoughts before each utterance, we capture both first- and second-order mental states across five categories: belief, intention, desire, emotion, and knowledge. These verbalized thoughts serve as answers to questions designed to assess the mental states of characters within conversations. Furthermore, the information asymmetry introduced by hiding thoughts from others induces the generation of false beliefs about various mental states. Assigning distinct personality traits to LLMs further diversifies both utterances and thoughts. ToMATO consists of 5.4k questions, 753 conversations, and 15 personality trait patterns. Our analysis shows that this dataset construction approach frequently generates false beliefs due to the information asymmetry between role-playing LLMs, and effectively reflects diverse personalities. We evaluate nine LLMs on ToMATO and find that even GPT-4o mini lags behind human performance, especially in understanding false beliefs, and lacks robustness to various personality traits.
comment: Accepted by AAAI 2025
☆ MMDocIR: Benchmarking Multi-Modal Retrieval for Long Documents
Multi-modal document retrieval is designed to identify and retrieve various forms of multi-modal content, such as figures, tables, charts, and layout information from extensive documents. Despite its significance, there is a notable lack of a robust benchmark to effectively evaluate the performance of systems in multi-modal document retrieval. To address this gap, this work introduces a new benchmark, named as MMDocIR, encompassing two distinct tasks: page-level and layout-level retrieval. The former focuses on localizing the most relevant pages within a long document, while the latter targets the detection of specific layouts, offering a more fine-grained granularity than whole-page analysis. A layout can refer to a variety of elements such as textual paragraphs, equations, figures, tables, or charts. The MMDocIR benchmark comprises a rich dataset featuring expertly annotated labels for 1,685 questions and bootstrapped labels for 173,843 questions, making it a pivotal resource for advancing multi-modal document retrieval for both training and evaluation. Through rigorous experiments, we reveal that (i) visual retrievers significantly outperform their text counterparts, (ii) MMDocIR train set can effectively benefit the training process of multi-modal document retrieval and (iii) text retrievers leveraging on VLM-text perform much better than those using OCR-text. These findings underscores the potential advantages of integrating visual elements for multi-modal document retrieval.
comment: https://huggingface.co/MMDocIR
☆ IDEA: Image Description Enhanced CLIP-Adapter
CLIP (Contrastive Language-Image Pre-training) has attained great success in pattern recognition and computer vision. Transferring CLIP to downstream tasks (e.g. zero- or few-shot classification) is a hot topic in multimodal learning. However, current studies primarily focus on either prompt learning for text or adapter tuning for vision, without fully exploiting the complementary information and correlations among image-text pairs. In this paper, we propose an Image Description Enhanced CLIP-Adapter (IDEA) method to adapt CLIP to few-shot image classification tasks. This method captures fine-grained features by leveraging both visual features and textual descriptions of images. IDEA is a training-free method for CLIP, and it can be comparable to or even exceeds state-of-the-art models on multiple tasks. Furthermore, we introduce Trainable-IDEA (T-IDEA), which extends IDEA by adding two lightweight learnable components (i.e., a projector and a learnable latent space), further enhancing the model's performance and achieving SOTA results on 11 datasets. As one important contribution, we employ the Llama model and design a comprehensive pipeline to generate textual descriptions for images of 11 datasets, resulting in a total of 1,637,795 image-text pairs, named "IMD-11". Our code and data are released at https://github.com/FourierAI/IDEA.
☆ SAIF: A Comprehensive Framework for Evaluating the Risks of Generative AI in the Public Sector AAAI
The rapid adoption of generative AI in the public sector, encompassing diverse applications ranging from automated public assistance to welfare services and immigration processes, highlights its transformative potential while underscoring the pressing need for thorough risk assessments. Despite its growing presence, evaluations of risks associated with AI-driven systems in the public sector remain insufficiently explored. Building upon an established taxonomy of AI risks derived from diverse government policies and corporate guidelines, we investigate the critical risks posed by generative AI in the public sector while extending the scope to account for its multimodal capabilities. In addition, we propose a Systematic dAta generatIon Framework for evaluating the risks of generative AI (SAIF). SAIF involves four key stages: breaking down risks, designing scenarios, applying jailbreak methods, and exploring prompt types. It ensures the systematic and consistent generation of prompt data, facilitating a comprehensive evaluation while providing a solid foundation for mitigating the risks. Furthermore, SAIF is designed to accommodate emerging jailbreak methods and evolving prompt types, thereby enabling effective responses to unforeseen risk scenarios. We believe that this study can play a crucial role in fostering the safe and responsible integration of generative AI into the public sector.
comment: 6 pages, 2 figures, 1 tables. AI for Public Missions (AIPM) Workshop at the 39th AAAI Conference on Artificial Intelligence (AAAI 2025)
☆ XMusic: Towards a Generalized and Controllable Symbolic Music Generation Framework
In recent years, remarkable advancements in artificial intelligence-generated content (AIGC) have been achieved in the fields of image synthesis and text generation, generating content comparable to that produced by humans. However, the quality of AI-generated music has not yet reached this standard, primarily due to the challenge of effectively controlling musical emotions and ensuring high-quality outputs. This paper presents a generalized symbolic music generation framework, XMusic, which supports flexible prompts (i.e., images, videos, texts, tags, and humming) to generate emotionally controllable and high-quality symbolic music. XMusic consists of two core components, XProjector and XComposer. XProjector parses the prompts of various modalities into symbolic music elements (i.e., emotions, genres, rhythms and notes) within the projection space to generate matching music. XComposer contains a Generator and a Selector. The Generator generates emotionally controllable and melodious music based on our innovative symbolic music representation, whereas the Selector identifies high-quality symbolic music by constructing a multi-task learning scheme involving quality assessment, emotion recognition, and genre recognition tasks. In addition, we build XMIDI, a large-scale symbolic music dataset that contains 108,023 MIDI files annotated with precise emotion and genre labels. Objective and subjective evaluations show that XMusic significantly outperforms the current state-of-the-art methods with impressive music quality. Our XMusic has been awarded as one of the nine Highlights of Collectibles at WAIC 2023. The project homepage of XMusic is https://xmusic-project.github.io.
comment: accepted by TMM
☆ Networked Agents in the Dark: Team Value Learning under Partial Observability AAMAS 2025
We propose a novel cooperative multi-agent reinforcement learning (MARL) approach for networked agents. In contrast to previous methods that rely on complete state information or joint observations, our agents must learn how to reach shared objectives under partial observability. During training, they collect individual rewards and approximate a team value function through local communication, resulting in cooperative behavior. To describe our problem, we introduce the networked dynamic partially observable Markov game framework, where agents communicate over a switching topology communication network. Our distributed method, DNA-MARL, uses a consensus mechanism for local communication and gradient descent for local computation. DNA-MARL increases the range of the possible applications of networked agents, being well-suited for real world domains that impose privacy and where the messages may not reach their recipients. We evaluate DNA-MARL across benchmark MARL scenarios. Our results highlight the superior performance of DNA-MARL over previous methods.
comment: 18 pages, 7 figures, 5 tables. Accepted as supplemental material at Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2025), Detroit, Michigan, USA, May 19 - 23, 2025, IFAAMAS
☆ How Developers Interact with AI: A Taxonomy of Human-AI Collaboration in Software Engineering
Artificial intelligence (AI), including large language models and generative AI, is emerging as a significant force in software development, offering developers powerful tools that span the entire development lifecycle. Although software engineering research has extensively studied AI tools in software development, the specific types of interactions between developers and these AI-powered tools have only recently begun to receive attention. Understanding and improving these interactions has the potential to improve productivity, trust, and efficiency in AI-driven workflows. In this paper, we propose a taxonomy of interaction types between developers and AI tools, identifying eleven distinct interaction types, such as auto-complete code suggestions, command-driven actions, and conversational assistance. Building on this taxonomy, we outline a research agenda focused on optimizing AI interactions, improving developer control, and addressing trust and usability challenges in AI-assisted development. By establishing a structured foundation for studying developer-AI interactions, this paper aims to stimulate research on creating more effective, adaptive AI tools for software development.
comment: Accepted at 2nd ACM International Conference on AI Foundation Models and Software Engineering (FORGE 2025)
☆ Leveraging LLM Agents for Translating Network Configurations
Configuration translation is a critical and frequent task in network operations. When a network device is damaged or outdated, administrators need to replace it to maintain service continuity. The replacement devices may originate from different vendors, necessitating configuration translation to ensure seamless network operation. However, translating configurations manually is a labor-intensive and error-prone process. In this paper, we propose an intent-based framework for translating network configuration with Large Language Model (LLM) Agents. The core of our approach is an Intent-based Retrieval Augmented Generation (IRAG) module that systematically splits a configuration file into fragments, extracts intents, and generates accurate translations. We also design a two-stage verification method to validate the syntax and semantics correctness of the translated configurations. We implement and evaluate the proposed method on real-world network configurations. Experimental results show that our method achieves 97.74% syntax correctness, outperforming state-of-the-art methods in translation accuracy.
Self-supervised Transformation Learning for Equivariant Representations NeurIPS 2024
Unsupervised representation learning has significantly advanced various machine learning tasks. In the computer vision domain, state-of-the-art approaches utilize transformations like random crop and color jitter to achieve invariant representations, embedding semantically the same inputs despite transformations. However, this can degrade performance in tasks requiring precise features, such as localization or flower classification. To address this, recent research incorporates equivariant representation learning, which captures transformation-sensitive information. However, current methods depend on transformation labels and thus struggle with interdependency and complex transformations. We propose Self-supervised Transformation Learning (STL), replacing transformation labels with transformation representations derived from image pairs. The proposed method ensures transformation representation is image-invariant and learns corresponding equivariant transformations, enhancing performance without increased batch complexity. We demonstrate the approach's effectiveness across diverse classification and detection tasks, outperforming existing methods in 7 out of 11 benchmarks and excelling in detection. By integrating complex transformations like AugMix, unusable by prior equivariant methods, this approach enhances performance across tasks, underscoring its adaptability and resilience. Additionally, its compatibility with various base models highlights its flexibility and broad applicability. The code is available at https://github.com/jaemyung-u/stl.
comment: 38th Conference on Neural Information Processing Systems (NeurIPS 2024)
☆ SPEQ: Stabilization Phases for Efficient Q-Learning in High Update-To-Data Ratio Reinforcement Learning
A key challenge in Deep Reinforcement Learning is sample efficiency, especially in real-world applications where collecting environment interactions is expensive or risky. Recent off-policy algorithms improve sample efficiency by increasing the Update-To-Data (UTD) ratio and performing more gradient updates per environment interaction. While this improves sample efficiency, it significantly increases computational cost due to the higher number of gradient updates required. In this paper we propose a sample-efficient method to improve computational efficiency by separating training into distinct learning phases in order to exploit gradient updates more effectively. Our approach builds on top of the Dropout Q-Functions (DroQ) algorithm and alternates between an online, low UTD ratio training phase, and an offline stabilization phase. During the stabilization phase, we fine-tune the Q-functions without collecting new environment interactions. This process improves the effectiveness of the replay buffer and reduces computational overhead. Our experimental results on continuous control problems show that our method achieves results comparable to state-of-the-art, high UTD ratio algorithms while requiring 56\% fewer gradient updates and 50\% less training time than DroQ. Our approach offers an effective and computationally economical solution while maintaining the same sample efficiency as the more costly, high UTD ratio state-of-the-art.
☆ Application of Deep Reinforcement Learning to UAV Swarming for Ground Surveillance
This paper summarizes in depth the state of the art of aerial swarms, covering both classical and new reinforcement-learning-based approaches for their management. Then, it proposes a hybrid AI system, integrating deep reinforcement learning in a multi-agent centralized swarm architecture. The proposed system is tailored to perform surveillance of a specific area, searching and tracking ground targets, for security and law enforcement applications. The swarm is governed by a central swarm controller responsible for distributing different search and tracking tasks among the cooperating UAVs. Each UAV agent is then controlled by a collection of cooperative sub-agents, whose behaviors have been trained using different deep reinforcement learning models, tailored for the different task types proposed by the swarm controller. More specifically, proximal policy optimization (PPO) algorithms were used to train the agents' behavior. In addition, several metrics to assess the performance of the swarm in this application were defined. The results obtained through simulation show that our system searches the operation area effectively, acquires the targets in a reasonable time, and is capable of tracking them continuously and consistently.
☆ Fine-grained Spatio-temporal Event Prediction with Self-adaptive Anchor Graph SDM'25
Event prediction tasks often handle spatio-temporal data distributed in a large spatial area. Different regions in the area exhibit different characteristics while having latent correlations. This spatial heterogeneity and correlations greatly affect the spatio-temporal distributions of event occurrences, which has not been addressed by state-of-the-art models. Learning spatial dependencies of events in a continuous space is challenging due to its fine granularity and a lack of prior knowledge. In this work, we propose a novel Graph Spatio-Temporal Point Process (GSTPP) model for fine-grained event prediction. It adopts an encoder-decoder architecture that jointly models the state dynamics of spatially localized regions using neural Ordinary Differential Equations (ODEs). The state evolution is built on the foundation of a novel Self-Adaptive Anchor Graph (SAAG) that captures spatial dependencies. By adaptively localizing the anchor nodes in the space and jointly constructing the correlation edges between them, the SAAG enhances the model's ability of learning complex spatial event patterns. The proposed GSTPP model greatly improves the accuracy of fine-grained event prediction. Extensive experimental results show that our method greatly improves the prediction accuracy over existing spatio-temporal event prediction approaches.
comment: Accepted to SIAM International Conference on Data Mining 2025 (SDM'25)
☆ MAGNET: Augmenting Generative Decoders with Representation Learning and Infilling Capabilities
While originally designed for unidirectional generative modeling, decoder-only large language models (LLMs) are increasingly being adapted for bidirectional modeling. However, unidirectional and bidirectional models are typically trained separately with distinct objectives (generation and representation learning, respectively). This separation overlooks the opportunity for developing a more versatile language model and for these objectives to complement each other. In this work, we introduce MAGNET, an adaptation of decoder-only LLMs that enhances their ability to generate robust representations and infill missing text spans, while preserving their knowledge and text generation capabilities. MAGNET employs three self-supervised training objectives and introduces an attention mechanism that combines bidirectional and causal attention, enabling unified training across all objectives. Our results demonstrate that LLMs adapted with MAGNET (1) surpass strong text encoders on token-level and sentence-level representation learning tasks, (2) generate contextually appropriate text infills by leveraging future context, (3) retain the ability for open-ended text generation without exhibiting repetition problem, and (4) preserve the knowledge gained by the LLM during pretraining.
☆ Reassessing the Role of Chain-of-Thought in Sentiment Analysis: Insights and Limitations
The relationship between language and thought remains an unresolved philosophical issue. Existing viewpoints can be broadly categorized into two schools: one asserting their independence, and another arguing that language constrains thought. In the context of large language models, this debate raises a crucial question: Does a language model's grasp of semantic meaning depend on thought processes? To explore this issue, we investigate whether reasoning techniques can facilitate semantic understanding. Specifically, we conceptualize thought as reasoning, employ chain-of-thought prompting as a reasoning technique, and examine its impact on sentiment analysis tasks. The experiments show that chain-of-thought has a minimal impact on sentiment analysis tasks. Both the standard and chain-of-thought prompts focus on aspect terms rather than sentiment in the generated content. Furthermore, counterfactual experiments reveal that the model's handling of sentiment tasks primarily depends on information from demonstrations. The experimental results support the first viewpoint.
☆ ViBidirectionMT-Eval: Machine Translation for Vietnamese-Chinese and Vietnamese-Lao language pair
This paper presents an results of the VLSP 2022-2023 Machine Translation Shared Tasks, focusing on Vietnamese-Chinese and Vietnamese-Lao machine translation. The tasks were organized as part of the 9th, 10th annual workshop on Vietnamese Language and Speech Processing (VLSP 2022, VLSP 2023). The objective of the shared task was to build machine translation systems, specifically targeting Vietnamese-Chinese and Vietnamese-Lao translation (corresponding to 4 translation directions). The submission were evaluated on 1,000 pairs for testing (news and general domains) using established metrics like BLEU [11] and SacreBLEU [12]. Additionally, system outputs also were evaluated with human judgment provided by experts in Chinese and Lao languages. These human assessments played a crucial role in ranking the performance of the machine translation models, ensuring a more comprehensive evaluation.
☆ Disjoint Processing Mechanisms of Hierarchical and Linear Grammars in Large Language Models
All natural languages are structured hierarchically. In humans, this structural restriction is neurologically coded: when two grammars are presented with identical vocabularies, brain areas responsible for language processing are only sensitive to hierarchical grammars. Using large language models (LLMs), we investigate whether such functionally distinct hierarchical processing regions can arise solely from exposure to large-scale language distributions. We generate inputs using English, Italian, Japanese, or nonce words, varying the underlying grammars to conform to either hierarchical or linear/positional rules. Using these grammars, we first observe that language models show distinct behaviors on hierarchical versus linearly structured inputs. Then, we find that the components responsible for processing hierarchical grammars are distinct from those that process linear grammars; we causally verify this in ablation experiments. Finally, we observe that hierarchy-selective components are also active on nonce grammars; this suggests that hierarchy sensitivity is not tied to meaning, nor in-distribution inputs.
☆ RLHS: Mitigating Misalignment in RLHF with Hindsight Simulation
Generative AI systems like foundation models (FMs) must align well with human values to ensure their behavior is helpful and trustworthy. While Reinforcement Learning from Human Feedback (RLHF) has shown promise for optimizing model performance using human judgments, existing RLHF pipelines predominantly rely on immediate feedback, which can fail to accurately reflect the downstream impact of an interaction on users' utility. We demonstrate that feedback based on evaluators' foresight estimates of downstream consequences systematically induces Goodhart's Law dynamics, incentivizing misaligned behaviors like sycophancy and deception and ultimately degrading user outcomes. To alleviate this, we propose decoupling evaluation from prediction by refocusing RLHF on hindsight feedback. Our theoretical analysis reveals that conditioning evaluator feedback on downstream observations mitigates misalignment and improves expected human utility, even when these observations are simulated by the AI system itself. To leverage this insight in a practical alignment algorithm, we introduce Reinforcement Learning from Hindsight Simulation (RLHS), which first simulates plausible consequences and then elicits feedback to assess what behaviors were genuinely beneficial in hindsight. We apply RLHS to two widely-employed online and offline preference optimization methods -- Proximal Policy Optimization (PPO) and Direct Preference Optimization (DPO) -- and show empirically that misalignment is significantly reduced with both methods. Through an online human user study, we show that RLHS consistently outperforms RLHF in helping users achieve their goals and earns higher satisfaction ratings, despite being trained solely with simulated hindsight feedback. These results underscore the importance of focusing on long-term consequences, even simulated ones, to mitigate misalignment in RLHF.
☆ Monte Carlo Tree Search for Comprehensive Exploration in LLM-Based Automatic Heuristic Design
Handcrafting heuristics for solving complex planning tasks (e.g., NP-hard combinatorial optimization (CO) problems) is a common practice but requires extensive domain knowledge. Recently, Large Language Model (LLM)-based automatic heuristics design (AHD) methods have shown promise in generating high-quality heuristics without manual intervention. Existing LLM-based AHD methods employ a population to maintain a fixed number of top-performing LLM-generated heuristics and introduce evolutionary computation (EC) to enhance the population iteratively. However, the population-based procedure brings greedy properties, often resulting in convergence to local optima. Instead, to more comprehensively explore the space of heuristics, we propose using Monte Carlo Tree Search (MCTS) for LLM-based heuristic evolution while preserving all LLM-generated heuristics in a tree structure. With a novel thought-alignment process and an exploration-decay technique, the proposed MCTS-AHD method delivers significantly higher-quality heuristics on various complex tasks. Our code is available at https://github.com/zz1358m/MCTS-AHD-master.
☆ AutoRestTest: A Tool for Automated REST API Testing Using LLMs and MARL ICSE
As REST APIs have become widespread in modern web services, comprehensive testing of these APIs has become increasingly crucial. Due to the vast search space consisting of operations, parameters, and parameter values along with their complex dependencies and constraints, current testing tools suffer from low code coverage, leading to suboptimal fault detection. To address this limitation, we present a novel tool, AutoRestTest, which integrates the Semantic Operation Dependency Graph (SODG) with Multi-Agent Reinforcement Learning (MARL) and large language models (LLMs) for effective REST API testing. AutoRestTest determines operation-dependent parameters using the SODG and employs five specialized agents (operation, parameter, value, dependency, and header) to identify dependencies of operations and generate operation sequences, parameter combinations, and values. AutoRestTest provides a command-line interface and continuous telemetry on successful operation count, unique server errors detected, and time elapsed. Upon completion, AutoRestTest generates a detailed report highlighting errors detected and operations exercised. In this paper, we introduce our tool and present preliminary results.
comment: To be published in the 47th IEEE/ACM International Conference on Software Engineering - Demonstration Track (ICSE-Demo 2025)
☆ LlamaRestTest: Effective REST API Testing with Small Language Models
Modern web services rely heavily on REST APIs, typically documented using the OpenAPI specification. The widespread adoption of this standard has resulted in the development of many black-box testing tools that generate tests based on these specifications. Recent advancements in Natural Language Processing (NLP), particularly with Large Language Models (LLMs), have enhanced REST API testing by extracting actionable rules and generating input values from the human-readable portions of the specification. However, these advancements overlook the potential of continuously refining the identified rules and test inputs based on server responses. To address this limitation, we present LlamaRestTest, a novel approach that employs two custom LLMs to generate realistic test inputs and uncover parameter dependencies during the testing process by incorporating server responses. These LLMs are created by fine-tuning the Llama3-8b model, using mined datasets of REST API example values and inter-parameter dependencies. We evaluated LlamaRestTest on 12 real-world services (including popular services such as Spotify), comparing it against RESTGPT, a GPT-powered specification-enhancement tool, as well as several state-of-the-art REST API testing tools, including RESTler, MoRest, EvoMaster, and ARAT-RL. Our results show that fine-tuning enables smaller LLMs to outperform larger models in detecting actionable rules and generating inputs for REST API testing. We evaluated configurations from the base Llama3-8B to fine-tuned versions and explored 2-bit, 4-bit, and 8-bit quantization for efficiency. LlamaRestTest surpasses state-of-the-art tools in code coverage and error detection, even with RESTGPT-enhanced specifications, and an ablation study highlights the impact of its novel components.
comment: To be published in the ACM International Conference on the Foundations of Software Engineering (FSE 2025)
☆ OpenMLDB: A Real-Time Relational Data Feature Computation System for Online ML
Efficient and consistent feature computation is crucial for a wide range of online ML applications. Typically, feature computation is divided into two distinct phases, i.e., offline stage for model training and online stage for model serving. These phases often rely on execution engines with different interface languages and function implementations, causing significant inconsistencies. Moreover, many online ML features involve complex time-series computations (e.g., functions over varied-length table windows) that differ from standard streaming and analytical queries. Existing data processing systems (e.g., Spark, Flink, DuckDB) often incur multi-second latencies for these computations, making them unsuitable for real-time online ML applications that demand timely feature updates. This paper presents OpenMLDB, a feature computation system deployed in 4Paradigm's SageOne platform and over 100 real scenarios. Technically, OpenMLDB first employs a unified query plan generator for consistent computation results across the offline and online stages, significantly reducing feature deployment overhead. Second, OpenMLDB provides an online execution engine that resolves performance bottlenecks caused by long window computations (via pre-aggregation) and multi-table window unions (via data self-adjusting). It also provides a high-performance offline execution engine with window parallel optimization and time-aware data skew resolving. Third, OpenMLDB features a compact data format and stream-focused indexing to maximize memory usage and accelerate data access. Evaluations in testing and real workloads reveal significant performance improvements and resource savings compared to the baseline systems. The open community of OpenMLDB now has over 150 contributors and gained 1.6k stars on GitHub.
☆ Sound Scene Synthesis at the DCASE 2024 Challenge
This paper presents Task 7 at the DCASE 2024 Challenge: sound scene synthesis. Recent advances in sound synthesis and generative models have enabled the creation of realistic and diverse audio content. We introduce a standardized evaluation framework for comparing different sound scene synthesis systems, incorporating both objective and subjective metrics. The challenge attracted four submissions, which are evaluated using the Fr\'echet Audio Distance (FAD) and human perceptual ratings. Our analysis reveals significant insights into the current capabilities and limitations of sound scene synthesis systems, while also highlighting areas for future improvement in this rapidly evolving field.
☆ Evaluating SAT and SMT Solvers on Large-Scale Sudoku Puzzles
Modern SMT solvers have revolutionized the approach to constraint satisfaction problems by integrating advanced theory reasoning and encoding techniques. In this work, we evaluate the performance of modern SMT solvers in Z3, CVC5 and DPLL(T) against a standard SAT solver in DPLL. By benchmarking these solvers on novel, diverse 25x25 Sudoku puzzles of various difficulty levels created by our improved Sudoku generator, we examine the impact of advanced theory reasoning and encoding techniques. Our findings demonstrate that modern SMT solvers significantly outperform classical SAT solvers. This work highlights the evolution of logical solvers and exemplifies the utility of SMT solvers in addressing large-scale constraint satisfaction problems.
☆ Towards Lightweight and Stable Zero-shot TTS with Self-distilled Representation Disentanglement
Zero-shot Text-To-Speech (TTS) synthesis shows great promise for personalized voice customization through voice cloning. However, current methods for achieving zero-shot TTS heavily rely on large model scales and extensive training datasets to ensure satisfactory performance and generalizability across various speakers. This raises concerns regarding both deployment costs and data security. In this paper, we present a lightweight and stable zero-shot TTS system. We introduce a novel TTS architecture designed to effectively model linguistic content and various speaker attributes from source speech and prompt speech, respectively. Furthermore, we present a two-stage self-distillation framework that constructs parallel data pairs for effectively disentangling linguistic content and speakers from the perspective of training data. Extensive experiments show that our system exhibits excellent performance and superior stability on the zero-shot TTS tasks. Moreover, it shows markedly superior computational efficiency, with RTFs of 0.13 and 0.012 on the CPU and GPU, respectively.
comment: 5 pages,4 figures
☆ DualOpt: A Dual Divide-and-Optimize Algorithm for the Large-scale Traveling Salesman Problem AAAI-25
This paper proposes a dual divide-and-optimize algorithm (DualOpt) for solving the large-scale traveling salesman problem (TSP). DualOpt combines two complementary strategies to improve both solution quality and computational efficiency. The first strategy is a grid-based divide-and-conquer procedure that partitions the TSP into smaller sub-problems, solving them in parallel and iteratively refining the solution by merging nodes and partial routes. The process continues until only one grid remains, yielding a high-quality initial solution. The second strategy involves a path-based divide-and-optimize procedure that further optimizes the solution by dividing it into sub-paths, optimizing each using a neural solver, and merging them back to progressively improve the overall solution. Extensive experiments conducted on two groups of TSP benchmark instances, including randomly generated instances with up to 100,000 nodes and real-world datasets from TSPLIB, demonstrate the effectiveness of DualOpt. The proposed DualOpt achieves highly competitive results compared to 10 state-of-the-art algorithms in the literature. In particular, DualOpt achieves an improvement gap up to 1.40% for the largest instance TSP100K with a remarkable 104x speed-up over the leading heuristic solver LKH3. Additionally, DualOpt demonstrates strong generalization on TSPLIB benchmarks, confirming its capability to tackle diverse real-world TSP applications.
comment: Accepted by AAAI-25, February 2025
☆ ANSR-DT: An Adaptive Neuro-Symbolic Learning and Reasoning Framework for Digital Twins
In this paper, we propose an Adaptive Neuro-Symbolic Learning Framework for digital twin technology called ``ANSR-DT." Our approach combines pattern recognition algorithms with reinforcement learning and symbolic reasoning to enable real-time learning and adaptive intelligence. This integration enhances the understanding of the environment and promotes continuous learning, leading to better and more effective decision-making in real-time for applications that require human-machine collaboration. We evaluated the \textit{ANSR-DT} framework for its ability to learn and adapt to dynamic patterns, observing significant improvements in decision accuracy, reliability, and interpretability when compared to existing state-of-the-art methods. However, challenges still exist in extracting and integrating symbolic rules in complex environments, which limits the full potential of our framework in heterogeneous settings. Moreover, our ongoing research aims to address this issue in the future by ensuring seamless integration of neural models at large. In addition, our open-source implementation promotes reproducibility and encourages future research to build on our foundational work.
☆ LAMS: LLM-Driven Automatic Mode Switching for Assistive Teleoperation
Teleoperating high degrees-of-freedom (DoF) robotic manipulators via low-DoF controllers like joysticks often requires frequent switching between control modes, where each mode maps controller movements to specific robot actions. Manually performing this frequent switching can make teleoperation cumbersome and inefficient. On the other hand, existing automatic mode-switching solutions, such as heuristic-based or learning-based methods, are often task-specific and lack generalizability. In this paper, we introduce LLM-Driven Automatic Mode Switching (LAMS), a novel approach that leverages Large Language Models (LLMs) to automatically switch control modes based on task context. Unlike existing methods, LAMS requires no prior task demonstrations and incrementally improves by integrating user-generated mode-switching examples. We validate LAMS through an ablation study and a user study with 10 participants on complex, long-horizon tasks, demonstrating that LAMS effectively reduces manual mode switches, is preferred over alternative methods, and improves performance over time. The project website with supplementary materials is at https://lams-assistance.github.io/.
☆ Reinforcement Learning-Enhanced Procedural Generation for Dynamic Narrative-Driven AR Experiences
Procedural Content Generation (PCG) is widely used to create scalable and diverse environments in games. However, existing methods, such as the Wave Function Collapse (WFC) algorithm, are often limited to static scenarios and lack the adaptability required for dynamic, narrative-driven applications, particularly in augmented reality (AR) games. This paper presents a reinforcement learning-enhanced WFC framework designed for mobile AR environments. By integrating environment-specific rules and dynamic tile weight adjustments informed by reinforcement learning (RL), the proposed method generates maps that are both contextually coherent and responsive to gameplay needs. Comparative evaluations and user studies demonstrate that the framework achieves superior map quality and delivers immersive experiences, making it well-suited for narrative-driven AR games. Additionally, the method holds promise for broader applications in education, simulation training, and immersive extended reality (XR) experiences, where dynamic and adaptive environments are critical.
comment: Number of pages: 13, Number of figures: 4. Accepted for presentation at GRAPP 2025 - 20th International Conference on Computer Graphics Theory and Applications (for additional details on the conference visit https://grapp.scitevents.org). Disclaimer: This preprint may differ from the final version published in the conference proceedings
☆ The Devil is in Temporal Token: High Quality Video Reasoning Segmentation
Existing methods for Video Reasoning Segmentation rely heavily on a single special token to represent the object in the keyframe or the entire video, inadequately capturing spatial complexity and inter-frame motion. To overcome these challenges, we propose VRS-HQ, an end-to-end video reasoning segmentation approach that leverages Multimodal Large Language Models (MLLMs) to inject rich spatiotemporal features into hierarchical tokens.Our key innovations include a Temporal Dynamic Aggregation (TDA) and a Token-driven Keyframe Selection (TKS). Specifically, we design frame-level and temporal-level tokens that utilize MLLM's autoregressive learning to effectively capture both local and global information. Subsequently, we apply a similarity-based weighted fusion and frame selection strategy, then utilize SAM2 to perform keyframe segmentation and propagation. To enhance keyframe localization accuracy, the TKS filters keyframes based on SAM2's occlusion scores during inference. VRS-HQ achieves state-of-the-art performance on ReVOS, surpassing VISA by 5.9%/12.5%/9.1% in J&F scores across the three subsets. These results highlight the strong temporal reasoning and segmentation capabilities of our method. Code and model weights will be released at VRS-HQ.
☆ Knowledge prompt chaining for semantic modeling
The task of building semantics for structured data such as CSV, JSON, and XML files is highly relevant in the knowledge representation field. Even though we have a vast of structured data on the internet, mapping them to domain ontologies to build semantics for them is still very challenging as it requires the construction model to understand and learn graph-structured knowledge. Otherwise, the task will require human beings' effort and cost. In this paper, we proposed a novel automatic semantic modeling framework: Knowledge Prompt Chaining. It can serialize the graph-structured knowledge and inject it into the LLMs properly in a Prompt Chaining architecture. Through this knowledge injection and prompting chaining, the model in our framework can learn the structure information and latent space of the graph and generate the semantic labels and semantic graphs following the chains' insturction naturally. Based on experimental results, our method achieves better performance than existing leading techniques, despite using reduced structured input data.
☆ Dynamic Portfolio Optimization via Augmented DDPG with Quantum Price Levels-Based Trading Strategy
With the development of deep learning, Dynamic Portfolio Optimization (DPO) problem has received a lot of attention in recent years, not only in the field of finance but also in the field of deep learning. Some advanced research in recent years has proposed the application of Deep Reinforcement Learning (DRL) to the DPO problem, which demonstrated to be more advantageous than supervised learning in solving the DPO problem. However, there are still certain unsolved issues: 1) DRL algorithms usually have the problems of slow learning speed and high sample complexity, which is especially problematic when dealing with complex financial data. 2) researchers use DRL simply for the purpose of obtaining high returns, but pay little attention to the problem of risk control and trading strategy, which will affect the stability of model returns. In order to address these issues, in this study we revamped the intrinsic structure of the model based on the Deep Deterministic Policy Gradient (DDPG) and proposed the Augmented DDPG model. Besides, we also proposed an innovative risk control strategy based on Quantum Price Levels (QPLs) derived from Quantum Finance Theory (QFT). Our experimental results revealed that our model has better profitability as well as risk control ability with less sample complexity in the DPO problem compared to the baseline models.
comment: 8 pages
☆ Doc-Guided Sent2Sent++: A Sent2Sent++ Agent with Doc-Guided memory for Document-level Machine Translation
The field of artificial intelligence has witnessed significant advancements in natural language processing, largely attributed to the capabilities of Large Language Models (LLMs). These models form the backbone of Agents designed to address long-context dependencies, particularly in Document-level Machine Translation (DocMT). DocMT presents unique challenges, with quality, consistency, and fluency being the key metrics for evaluation. Existing approaches, such as Doc2Doc and Doc2Sent, either omit sentences or compromise fluency. This paper introduces Doc-Guided Sent2Sent++, an Agent that employs an incremental sentence-level forced decoding strategy \textbf{to ensure every sentence is translated while enhancing the fluency of adjacent sentences.} Our Agent leverages a Doc-Guided Memory, focusing solely on the summary and its translation, which we find to be an efficient approach to maintaining consistency. Through extensive testing across multiple languages and domains, we demonstrate that Sent2Sent++ outperforms other methods in terms of quality, consistency, and fluency. The results indicate that, our approach has achieved significant improvements in metrics such as s-COMET, d-COMET, LTCR-$1_f$, and document-level perplexity (d-ppl). The contributions of this paper include a detailed analysis of current DocMT research, the introduction of the Sent2Sent++ decoding method, the Doc-Guided Memory mechanism, and validation of its effectiveness across languages and domains.
☆ Mitigating Domain Shift in Federated Learning via Intra- and Inter-Domain Prototypes
Federated Learning (FL) has emerged as a decentralized machine learning technique, allowing clients to train a global model collaboratively without sharing private data. However, most FL studies ignore the crucial challenge of heterogeneous domains where each client has a distinct feature distribution, which is common in real-world scenarios. Prototype learning, which leverages the mean feature vectors within the same classes, has become a prominent solution for federated learning under domain skew. However, existing federated prototype learning methods only consider inter-domain prototypes on the server and overlook intra-domain characteristics. In this work, we introduce a novel federated prototype learning method, namely I$^2$PFL, which incorporates $\textbf{I}$ntra-domain and $\textbf{I}$nter-domain $\textbf{P}$rototypes, to mitigate domain shifts and learn a generalized global model across multiple domains in federated learning. To construct intra-domain prototypes, we propose feature alignment with MixUp-based augmented prototypes to capture the diversity of local domains and enhance the generalization of local features. Additionally, we introduce a reweighting mechanism for inter-domain prototypes to generate generalized prototypes to provide inter-domain knowledge and reduce domain skew across multiple clients. Extensive experiments on the Digits, Office-10, and PACS datasets illustrate the superior performance of our method compared to other baselines.
comment: 13 pages, 9 figures, 10 tables
☆ Easing Seasickness through Attention Redirection with a Mindfulness-Based Brain--Computer Interface
Seasickness is a prevalent issue that adversely impacts both passenger experiences and the operational efficiency of maritime crews. While techniques that redirect attention have proven effective in alleviating motion sickness symptoms in terrestrial environments, applying similar strategies to manage seasickness poses unique challenges due to the prolonged and intense motion environment associated with maritime travel. In this study, we propose a mindfulness brain-computer interface (BCI), specifically designed to redirect attention with the aim of mitigating seasickness symptoms in real-world settings. Our system utilizes a single-channel headband to capture prefrontal EEG signals, which are then wirelessly transmitted to computing devices for the assessment of mindfulness states. The results are transferred into real-time feedback as mindfulness scores and audiovisual stimuli, facilitating a shift in attentional focus from physiological discomfort to mindfulness practices. A total of 43 individuals participated in a real-world maritime experiment consisted of three sessions: a real-feedback mindfulness session, a resting session, and a pseudofeedback mindfulness session. Notably, 81.39% of participants reported that the mindfulness BCI intervention was effective, and there was a significant reduction in the severity of seasickness, as measured by the Misery Scale (MISC). Furthermore, EEG analysis revealed a decrease in the theta/beta ratio, corresponding with the alleviation of seasickness symptoms. A decrease in overall EEG band power during the real-feedback mindfulness session suggests that the mindfulness BCI fosters a more tranquil and downregulated state of brain activity. Together, this study presents a novel nonpharmacological, portable, and effective approach for seasickness intervention, with the potential to enhance the cruising experience for both passengers and crews.
☆ Exploring the Efficacy of Meta-Learning: Unveiling Superior Data Diversity Utilization of MAML Over Pre-training
Currently, data and model size dominate the narrative in the training of super-large, powerful models. However, there has been a lack of exploration on the effect of other attributes of the training dataset on model performance. We hypothesize that dataset diversity can impact the performance of vision models. Our study shows positive correlations between test set accuracy and data diversity, providing an argument for furthering the research of dataset attributes beyond size. We analyzed pre-training and model-agnostic meta-learning methods on twelve popular visual datasets (e.g., Omniglot, CIFAR-FS, Aircraft) and five model configurations, including MAML variants with different numbers of inner gradient steps and supervised learning. We show moderate to strong positive correlations (R-squared: 0.15-0.42) between accuracy and data diversity and weaker but significant correlations (R-squared: ~0.2) between loss and diversity. These findings support our hypothesis and demonstrate a promising way for a deeper exploration of how formal data diversity influences model performance. This initial study highlights the potential of (Task2Vec) data diversity as a valuable measure in the rapidly evolving field of large-scale learning and emphasizes that understanding the dataset is key to building more powerful and generalizable models.
☆ Adapting Whisper for Regional Dialects: Enhancing Public Services for Vulnerable Populations in the United Kingdom
We collect novel data in the public service domain to evaluate the capability of the state-of-the-art automatic speech recognition (ASR) models in capturing regional differences in accents in the United Kingdom (UK), specifically focusing on two accents from Scotland with distinct dialects. This study addresses real-world problems where biased ASR models can lead to miscommunication in public services, disadvantaging individuals with regional accents particularly those in vulnerable populations. We first examine the out-of-the-box performance of the Whisper large-v3 model on a baseline dataset and our data. We then explore the impact of fine-tuning Whisper on the performance in the two UK regions and investigate the effectiveness of existing model evaluation techniques for our real-world application through manual inspection of model errors. We observe that the Whisper model has a higher word error rate (WER) on our test datasets compared to the baseline data and fine-tuning on a given data improves performance on the test dataset with the same domain and accent. The fine-tuned models also appear to show improved performance when applied to the test data outside of the region it was trained on suggesting that fine-tuned models may be transferable within parts of the UK. Our manual analysis of model outputs reveals the benefits and drawbacks of using WER as an evaluation metric and fine-tuning to adapt to regional dialects.
☆ Grounding Text-To-Image Diffusion Models For Controlled High-Quality Image Generation
Large-scale text-to-image (T2I) diffusion models have demonstrated an outstanding performance in synthesizing diverse high-quality visuals from natural language text captions. Multiple layout-to-image models have been developed to control the generation process by utilizing a broad array of layouts such as segmentation maps, edges, and human keypoints. In this work, we present ObjectDiffusion, a model that takes inspirations from the top cutting-edge image generative frameworks to seamlessly condition T2I models with new bounding boxes capabilities. Specifically, we make substantial modifications to the network architecture introduced in ContorlNet to integrate it with the condition processing and injection techniques proposed in GLIGEN. ObjectDiffusion is initialized with pretraining parameters to leverage the generation knowledge obtained from training on large-scale datasets. We fine-tune ObjectDiffusion on the COCO2017 training dataset and evaluate it on the COCO2017 validation dataset. Our model achieves an AP$_{50}$ of 46.6, an AR of 44.5, and a FID of 19.8 outperforming the current SOTA model trained on open-source datasets in all of the three metrics. ObjectDiffusion demonstrates a distinctive capability in synthesizing diverse, high-quality, high-fidelity images that seamlessly conform to the semantic and spatial control layout. Evaluated in qualitative and quantitative tests, ObjectDiffusion exhibits remarkable grounding abilities on closed-set and open-set settings across a wide variety of contexts. The qualitative assessment verifies the ability of ObjectDiffusion to generate multiple objects of different sizes and locations.
☆ Patch-aware Vector Quantized Codebook Learning for Unsupervised Visual Defect Detection ICTAI 2024
Unsupervised visual defect detection is critical in industrial applications, requiring a representation space that captures normal data features while detecting deviations. Achieving a balance between expressiveness and compactness is challenging; an overly expressive space risks inefficiency and mode collapse, impairing detection accuracy. We propose a novel approach using an enhanced VQ-VAE framework optimized for unsupervised defect detection. Our model introduces a patch-aware dynamic code assignment scheme, enabling context-sensitive code allocation to optimize spatial representation. This strategy enhances normal-defect distinction and improves detection accuracy during inference. Experiments on MVTecAD, BTAD, and MTSD datasets show our method achieves state-of-the-art performance.
comment: 7 pages, Accepted to 36th IEEE ICTAI 2024
☆ Guiding Retrieval using LLM-based Listwise Rankers
Large Language Models (LLMs) have shown strong promise as rerankers, especially in ``listwise'' settings where an LLM is prompted to rerank several search results at once. However, this ``cascading'' retrieve-and-rerank approach is limited by the bounded recall problem: relevant documents not retrieved initially are permanently excluded from the final ranking. Adaptive retrieval techniques address this problem, but do not work with listwise rerankers because they assume a document's score is computed independently from other documents. In this paper, we propose an adaptation of an existing adaptive retrieval method that supports the listwise setting and helps guide the retrieval process itself (thereby overcoming the bounded recall problem for LLM rerankers). Specifically, our proposed algorithm merges results both from the initial ranking and feedback documents provided by the most relevant documents seen up to that point. Through extensive experiments across diverse LLM rerankers, first stage retrievers, and feedback sources, we demonstrate that our method can improve nDCG@10 by up to 13.23% and recall by 28.02%--all while keeping the total number of LLM inferences constant and overheads due to the adaptive process minimal. The work opens the door to leveraging LLM-based search in settings where the initial pool of results is limited, e.g., by legacy systems, or by the cost of deploying a semantic first-stage.
comment: 16 pages, 2 figures, 3 tables
☆ A Blockchain-Enabled Approach to Cross-Border Compliance and Trust
As artificial intelligence (AI) systems become increasingly integral to critical infrastructure and global operations, the need for a unified, trustworthy governance framework is more urgent that ever. This paper proposes a novel approach to AI governance, utilizing blockchain and distributed ledger technologies (DLT) to establish a decentralized, globally recognized framework that ensures security, privacy, and trustworthiness of AI systems across borders. The paper presents specific implementation scenarios within the financial sector, outlines a phased deployment timeline over the next decade, and addresses potential challenges with solutions grounded in current research. By synthesizing advancements in blockchain, AI ethics, and cybersecurity, this paper offers a comprehensive roadmap for a decentralized AI governance framework capable of adapting to the complex and evolving landscape of global AI regulation.
comment: This is a preprint of paper that has been accepted for Publication at 2024 IEEE International Conference on Trust, Privacy and Security in Intelligent Systems, and Applications
☆ Attention is All You Need Until You Need Retention
This work introduces a novel Retention Layer mechanism for Transformer based architectures, addressing their inherent lack of intrinsic retention capabilities. Unlike human cognition, which can encode and dynamically recall symbolic templates, Generative Pretrained Transformers rely solely on fixed pretrained weights and ephemeral context windows, limiting their adaptability. The proposed Retention Layer incorporates a persistent memory module capable of real time data population, dynamic recall, and guided output generation. This enhancement allows models to store, update, and reuse observed patterns across sessions, enabling incremental learning and bridging the gap between static pretraining and dynamic, context sensitive adaptation. The Retention Layer design parallels social learning processes, encompassing attention, retention, reproduction, and motivation stages. Technically, it integrates a memory attention mechanism and episodic buffers to manage memory scalability, mitigate overfitting, and ensure efficient recall. Applications span adaptive personal assistants, real time fraud detection, autonomous robotics, content moderation, and healthcare diagnostics. In each domain, the retention mechanism enables systems to learn incrementally, personalize outputs, and respond to evolving real world challenges effectively. By emulating key aspects of human learning, this retention enhanced architecture fosters a more fluid and responsive AI paradigm, paving the way for dynamic, session aware models that extend the capabilities of traditional Transformers into domains requiring continual adaptation.
☆ The Veln(ia)s is in the Details: Evaluating LLM Judgment on Latvian and Lithuanian Short Answer Matching
In this work, we address the challenge of evaluating large language models (LLMs) on the short answer matching task for Latvian and Lithuanian languages. We introduce novel datasets consisting of 502 Latvian and 690 Lithuanian question-answer pairs. For each question-answer pair, we generated matched and non-matched answers using a set of alteration rules specifically designed to introduce small but meaningful changes in the text. These generated answers serve as test cases to assess the ability of LLMs to detect subtle differences in matching of the original answers. A subset of the datasets was manually verified for quality and accuracy. Our results show that while larger LLMs, such as QWEN2.5 72b and LLaMa3.1 70b, demonstrate near-perfect performance in distinguishing matched and non-matched answers, smaller models show more variance. For instance, LLaMa3.1 8b and EuroLLM 9b benefited from few-shot examples, while Mistral Nemo 12b underperformed on detection of subtle text alteration, particularly in Lithuanian, even with additional examples. QWEN2.5 7b and Mistral 7b were able to obtain a strong and comparable performance to the larger 70b models in zero and few shot experiments. Moreover, the performance of Mistral 7b was weaker in few shot experiments.
☆ Towards Understanding Extrapolation: a Causal Lens NeurIPS 2024
Canonical work handling distribution shifts typically necessitates an entire target distribution that lands inside the training distribution. However, practical scenarios often involve only a handful of target samples, potentially lying outside the training support, which requires the capability of extrapolation. In this work, we aim to provide a theoretical understanding of when extrapolation is possible and offer principled methods to achieve it without requiring an on-support target distribution. To this end, we formulate the extrapolation problem with a latent-variable model that embodies the minimal change principle in causal mechanisms. Under this formulation, we cast the extrapolation problem into a latent-variable identification problem. We provide realistic conditions on shift properties and the estimation objectives that lead to identification even when only one off-support target sample is available, tackling the most challenging scenarios. Our theory reveals the intricate interplay between the underlying manifold's smoothness and the shift properties. We showcase how our theoretical results inform the design of practical adaptation algorithms. Through experiments on both synthetic and real-world data, we validate our theoretical findings and their practical implications.
comment: NeurIPS 2024
☆ AutoLoop: Fast Visual SLAM Fine-tuning through Agentic Curriculum Learning
Current visual SLAM systems face significant challenges in balancing computational efficiency with robust loop closure handling. Traditional approaches require careful manual tuning and incur substantial computational overhead, while learning-based methods either lack explicit loop closure capabilities or implement them through computationally expensive methods. We present AutoLoop, a novel approach that combines automated curriculum learning with efficient fine-tuning for visual SLAM systems. Our method employs a DDPG (Deep Deterministic Policy Gradient) agent to dynamically adjust loop closure weights during training, eliminating the need for manual hyperparameter search while significantly reducing the required training steps. The approach pre-computes potential loop closure pairs offline and leverages them through an agent-guided curriculum, allowing the model to adapt efficiently to new scenarios. Experiments conducted on TartanAir for training and validated across multiple benchmarks including KITTI, EuRoC, ICL-NUIM and TUM RGB-D demonstrate that AutoLoop achieves comparable or superior performance while reducing training time by an order of magnitude compared to traditional approaches. AutoLoop provides a practical solution for rapid adaptation of visual SLAM systems, automating the weight tuning process that traditionally requires multiple manual iterations. Our results show that this automated curriculum strategy not only accelerates training but also maintains or improves the model's performance across diverse environmental conditions.
☆ Towards Multilingual LLM Evaluation for Baltic and Nordic languages: A study on Lithuanian History
In this work, we evaluated Lithuanian and general history knowledge of multilingual Large Language Models (LLMs) on a multiple-choice question-answering task. The models were tested on a dataset of Lithuanian national and general history questions translated into Baltic, Nordic, and other languages (English, Ukrainian, Arabic) to assess the knowledge sharing from culturally and historically connected groups. We evaluated GPT-4o, LLaMa3.1 8b and 70b, QWEN2.5 7b and 72b, Mistral Nemo 12b, LLaMa3 8b, Mistral 7b, LLaMa3.2 3b, and Nordic fine-tuned models (GPT-SW3 and LLaMa3 8b). Our results show that GPT-4o consistently outperformed all other models across language groups, with slightly better results for Baltic and Nordic languages. Larger open-source models like QWEN2.5 72b and LLaMa3.1 70b performed well but showed weaker alignment with Baltic languages. Smaller models (Mistral Nemo 12b, LLaMa3.2 3b, QWEN 7B, LLaMa3.1 8B, and LLaMa3 8b) demonstrated gaps with LT-related alignment with Baltic languages while performing better on Nordic and other languages. The Nordic fine-tuned models did not surpass multilingual models, indicating that shared cultural or historical context alone does not guarantee better performance.
☆ Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG
Large Language Models (LLMs) have revolutionized artificial intelligence (AI) by enabling human like text generation and natural language understanding. However, their reliance on static training data limits their ability to respond to dynamic, real time queries, resulting in outdated or inaccurate outputs. Retrieval Augmented Generation (RAG) has emerged as a solution, enhancing LLMs by integrating real time data retrieval to provide contextually relevant and up-to-date responses. Despite its promise, traditional RAG systems are constrained by static workflows and lack the adaptability required for multistep reasoning and complex task management. Agentic Retrieval-Augmented Generation (Agentic RAG) transcends these limitations by embedding autonomous AI agents into the RAG pipeline. These agents leverage agentic design patterns reflection, planning, tool use, and multiagent collaboration to dynamically manage retrieval strategies, iteratively refine contextual understanding, and adapt workflows to meet complex task requirements. This integration enables Agentic RAG systems to deliver unparalleled flexibility, scalability, and context awareness across diverse applications. This survey provides a comprehensive exploration of Agentic RAG, beginning with its foundational principles and the evolution of RAG paradigms. It presents a detailed taxonomy of Agentic RAG architectures, highlights key applications in industries such as healthcare, finance, and education, and examines practical implementation strategies. Additionally, it addresses challenges in scaling these systems, ensuring ethical decision making, and optimizing performance for real-world applications, while providing detailed insights into frameworks and tools for implementing Agentic RAG
☆ Benchmarking Robustness of Contrastive Learning Models for Medical Image-Report Retrieval AAAI 2025
Medical images and reports offer invaluable insights into patient health. The heterogeneity and complexity of these data hinder effective analysis. To bridge this gap, we investigate contrastive learning models for cross-domain retrieval, which associates medical images with their corresponding clinical reports. This study benchmarks the robustness of four state-of-the-art contrastive learning models: CLIP, CXR-RePaiR, MedCLIP, and CXR-CLIP. We introduce an occlusion retrieval task to evaluate model performance under varying levels of image corruption. Our findings reveal that all evaluated models are highly sensitive to out-of-distribution data, as evidenced by the proportional decrease in performance with increasing occlusion levels. While MedCLIP exhibits slightly more robustness, its overall performance remains significantly behind CXR-CLIP and CXR-RePaiR. CLIP, trained on a general-purpose dataset, struggles with medical image-report retrieval, highlighting the importance of domain-specific training data. The evaluation of this work suggests that more effort needs to be spent on improving the robustness of these models. By addressing these limitations, we can develop more reliable cross-domain retrieval models for medical applications.
comment: This work is accepted to AAAI 2025 Workshop -- the 9th International Workshop on Health Intelligence
☆ Generative Medical Image Anonymization Based on Latent Code Projection and Optimization
Medical image anonymization aims to protect patient privacy by removing identifying information, while preserving the data utility to solve downstream tasks. In this paper, we address the medical image anonymization problem with a two-stage solution: latent code projection and optimization. In the projection stage, we design a streamlined encoder to project input images into a latent space and propose a co-training scheme to enhance the projection process. In the optimization stage, we refine the latent code using two deep loss functions designed to address the trade-off between identity protection and data utility dedicated to medical images. Through a comprehensive set of qualitative and quantitative experiments, we showcase the effectiveness of our approach on the MIMIC-CXR chest X-ray dataset by generating anonymized synthetic images that can serve as training set for detecting lung pathologies. Source codes are available at https://github.com/Huiyu-Li/GMIA.
comment: Conference
☆ Mantis Shrimp: Exploring Photometric Band Utilization in Computer Vision Networks for Photometric Redshift Estimation
We present Mantis Shrimp, a multi-survey deep learning model for photometric redshift estimation that fuses ultra-violet (GALEX), optical (PanSTARRS), and infrared (UnWISE) imagery. Machine learning is now an established approach for photometric redshift estimation, with generally acknowledged higher performance in areas with a high density of spectroscopically identified galaxies over template-based methods. Multiple works have shown that image-based convolutional neural networks can outperform tabular-based color/magnitude models. In comparison to tabular models, image models have additional design complexities: it is largely unknown how to fuse inputs from different instruments which have different resolutions or noise properties. The Mantis Shrimp model estimates the conditional density estimate of redshift using cutout images. The density estimates are well calibrated and the point estimates perform well in the distribution of available spectroscopically confirmed galaxies with (bias = 1e-2), scatter (NMAD = 2.44e-2) and catastrophic outlier rate ($\eta$=17.53$\%$). We find that early fusion approaches (e.g., resampling and stacking images from different instruments) match the performance of late fusion approaches (e.g., concatenating latent space representations), so that the design choice ultimately is left to the user. Finally, we study how the models learn to use information across bands, finding evidence that our models successfully incorporates information from all surveys. The applicability of our model to the analysis of large populations of galaxies is limited by the speed of downloading cutouts from external servers; however, our model could be useful in smaller studies such as generating priors over redshift for stellar population synthesis.
☆ A Non-autoregressive Model for Joint STT and TTS
In this paper, we take a step towards jointly modeling automatic speech recognition (STT) and speech synthesis (TTS) in a fully non-autoregressive way. We develop a novel multimodal framework capable of handling the speech and text modalities as input either individually or together. The proposed model can also be trained with unpaired speech or text data owing to its multimodal nature. We further propose an iterative refinement strategy to improve the STT and TTS performance of our model such that the partial hypothesis at the output can be fed back to the input of our model, thus iteratively improving both STT and TTS predictions. We show that our joint model can effectively perform both STT and TTS tasks, outperforming the STT-specific baseline in all tasks and performing competitively with the TTS-specific baseline across a wide range of evaluation metrics.
comment: 5 pages, 3 figures, 3 tables
☆ Tracking the Takes and Trajectories of English-Language News Narratives across Trustworthy and Worrisome Websites USENIX Security
Understanding how misleading and outright false information enters news ecosystems remains a difficult challenge that requires tracking how narratives spread across thousands of fringe and mainstream news websites. To do this, we introduce a system that utilizes encoder-based large language models and zero-shot stance detection to scalably identify and track news narratives and their attitudes across over 4,000 factually unreliable, mixed-reliability, and factually reliable English-language news websites. Running our system over an 18 month period, we track the spread of 146K news stories. Using network-based interference via the NETINF algorithm, we show that the paths of news narratives and the stances of websites toward particular entities can be used to uncover slanted propaganda networks (e.g., anti-vaccine and anti-Ukraine) and to identify the most influential websites in spreading these attitudes in the broader news ecosystem. We hope that increased visibility into our distributed news ecosystem can help with the reporting and fact-checking of propaganda and disinformation.
comment: To appear at USENIX Security Symposium 2025. Keywords: Misinformation, News, Narratives, LLMs, Stance-Detection
☆ SteLLA: A Structured Grading System Using LLMs with RAG
Large Language Models (LLMs) have shown strong general capabilities in many applications. However, how to make them reliable tools for some specific tasks such as automated short answer grading (ASAG) remains a challenge. We present SteLLA (Structured Grading System Using LLMs with RAG) in which a) Retrieval Augmented Generation (RAG) approach is used to empower LLMs specifically on the ASAG task by extracting structured information from the highly relevant and reliable external knowledge based on the instructor-provided reference answer and rubric, b) an LLM performs a structured and question-answering-based evaluation of student answers to provide analytical grades and feedback. A real-world dataset that contains students' answers in an exam was collected from a college-level Biology course. Experiments show that our proposed system can achieve substantial agreement with the human grader while providing break-down grades and feedback on all the knowledge points examined in the problem. A qualitative and error analysis of the feedback generated by GPT4 shows that GPT4 is good at capturing facts while may be prone to inferring too much implication from the given text in the grading task which provides insights into the usage of LLMs in the ASAG system.
☆ Inferring Transition Dynamics from Value Functions AAAI-25
In reinforcement learning, the value function is typically trained to solve the Bellman equation, which connects the current value to future values. This temporal dependency hints that the value function may contain implicit information about the environment's transition dynamics. By rearranging the Bellman equation, we show that a converged value function encodes a model of the underlying dynamics of the environment. We build on this insight to propose a simple method for inferring dynamics models directly from the value function, potentially mitigating the need for explicit model learning. Furthermore, we explore the challenges of next-state identifiability, discussing conditions under which the inferred dynamics model is well-defined. Our work provides a theoretical foundation for leveraging value functions in dynamics modeling and opens a new avenue for bridging model-free and model-based reinforcement learning.
comment: Accepted at the AAAI-25 8th Workshop on Generalization in Planning
☆ Average-Reward Reinforcement Learning with Entropy Regularization AAAI-25
The average-reward formulation of reinforcement learning (RL) has drawn increased interest in recent years due to its ability to solve temporally-extended problems without discounting. Independently, RL algorithms have benefited from entropy-regularization: an approach used to make the optimal policy stochastic, thereby more robust to noise. Despite the distinct benefits of the two approaches, the combination of entropy regularization with an average-reward objective is not well-studied in the literature and there has been limited development of algorithms for this setting. To address this gap in the field, we develop algorithms for solving entropy-regularized average-reward RL problems with function approximation. We experimentally validate our method, comparing it with existing algorithms on standard benchmarks for RL.
comment: Accepted at the AAAI-25 Eighth Workshop on Bridging the Gap Between AI Planning and Reinforcement Learning (PRL)
☆ Decompose-ToM: Enhancing Theory of Mind Reasoning in Large Language Models through Simulation and Task Decomposition COLING 2025
Theory of Mind (ToM) is the ability to understand and reflect on the mental states of others. Although this capability is crucial for human interaction, testing on Large Language Models (LLMs) reveals that they possess only a rudimentary understanding of it. Although the most capable closed-source LLMs have come close to human performance on some ToM tasks, they still perform poorly on complex variations of the task that involve more structured reasoning. In this work, we utilize the concept of "pretend-play", or ``Simulation Theory'' from cognitive psychology to propose ``Decompose-ToM'': an LLM-based inference algorithm that improves model performance on complex ToM tasks. We recursively simulate user perspectives and decompose the ToM task into a simpler set of functions: subject identification, question-reframing, world model updation, and knowledge availability. We test the algorithm on higher-order ToM tasks and a task testing for ToM capabilities in a conversational setting, demonstrating that our approach shows significant improvement across models compared to baseline methods while requiring minimal prompt tuning across tasks and no additional model training.
comment: Accepted to COLING 2025
♻ ☆ Reward Machines for Deep RL in Noisy and Uncertain Environments
Reward Machines provide an automaton-inspired structure for specifying instructions, safety constraints, and other temporally extended reward-worthy behaviour. By exposing the underlying structure of a reward function, they enable the decomposition of an RL task, leading to impressive gains in sample efficiency. Although Reward Machines and similar formal specifications have a rich history of application towards sequential decision-making problems, they critically rely on a ground-truth interpretation of the domain-specific vocabulary that forms the building blocks of the reward function--such ground-truth interpretations are elusive in the real world due in part to partial observability and noisy sensing. In this work, we explore the use of Reward Machines for Deep RL in noisy and uncertain environments. We characterize this problem as a POMDP and propose a suite of RL algorithms that exploit task structure under uncertain interpretation of the domain-specific vocabulary. Through theory and experiments, we expose pitfalls in naive approaches to this problem while simultaneously demonstrating how task structure can be successfully leveraged under noisy interpretations of the vocabulary.
♻ ☆ Consistency of Responses and Continuations Generated by Large Language Models on Social Media
Large Language Models (LLMs) demonstrate remarkable capabilities in text generation, yet their emotional consistency and semantic coherence in social media contexts remain insufficiently understood. This study investigates how LLMs handle emotional content and maintain semantic relationships through continuation and response tasks using two open-source models: Gemma and Llama. By analyzing climate change discussions from Twitter and Reddit, we examine emotional transitions, intensity patterns, and semantic similarity between human-authored and LLM-generated content. Our findings reveal that while both models maintain high semantic coherence, they exhibit distinct emotional patterns: Gemma shows a tendency toward negative emotion amplification, particularly anger, while maintaining certain positive emotions like optimism. Llama demonstrates superior emotional preservation across a broader spectrum of affects. Both models systematically generate responses with attenuated emotional intensity compared to human-authored content and show a bias toward positive emotions in response tasks. Additionally, both models maintain strong semantic similarity with original texts, though performance varies between continuation and response tasks. These findings provide insights into LLMs' emotional and semantic processing capabilities, with implications for their deployment in social media contexts and human-AI interaction design.
♻ ☆ Learning Low-Dimensional Strain Models of Soft Robots by Looking at the Evolution of Their Shape with Application to Model-Based Control
Obtaining dynamic models of continuum soft robots is central to the analysis and control of soft robots, and researchers have devoted much attention to the challenge of proposing both data-driven and first-principle solutions. Both avenues have, however, shown their limitations; the former lacks structure and performs poorly outside training data, while the latter requires significant simplifications and extensive expert knowledge to be used in practice. This paper introduces a streamlined method for learning low-dimensional, physics-based models that are both accurate and easy to interpret. We start with an algorithm that uses image data (i.e., shape evolutions) to determine the minimal necessary segments for describing a soft robot's movement. Following this, we apply a dynamic regression and strain sparsification algorithm to identify relevant strains and define the model's dynamics. We validate our approach through simulations with various planar soft manipulators, comparing its performance against other learning strategies, showing that our models are both computationally efficient and 25x more accurate on out-of-training distribution inputs. Finally, we demonstrate that thanks to the capability of the method of generating physically compatible models, the learned models can be straightforwardly combined with model-based control policies.
comment: 8 pages, appearing in Proceedings of the 2025 IEEE 8th International Conference on Soft Robotics (RoboSoft)
♻ ☆ A Discrete-sequence Dataset for Evaluating Online Unsupervised Anomaly Detection Approaches for Multivariate Time Series
Benchmarking anomaly detection approaches for multivariate time series is challenging due to the lack of high-quality datasets. Current publicly available datasets are too small, not diverse and feature trivial anomalies, which hinders measurable progress in this research area. We propose a solution: a diverse, extensive, and non-trivial dataset generated via state-of-the-art simulation tools that reflects realistic behaviour of an automotive powertrain, including its multivariate, dynamic and variable-state properties. To cater for both unsupervised and semi-supervised anomaly detection settings, as well as time series generation and forecasting, we make different versions of the dataset available, where training and test subsets are offered in contaminated and clean versions, depending on the task. We also provide baseline results from a small selection of approaches based on deterministic and variational autoencoders, as well as a non-parametric approach. As expected, the baseline experimentation shows that the approaches trained on the semi-supervised version of the dataset outperform their unsupervised counterparts, highlighting a need for approaches more robust to contaminated training data.
comment: Submitted to the IEEE Transactions on Reliability journal
♻ ☆ Identifying Spurious Correlations using Counterfactual Alignment
Models driven by spurious correlations often yield poor generalization performance. We propose the counterfactual (CF) alignment method to detect and quantify spurious correlations of black box classifiers. Our methodology is based on counterfactual images generated with respect to one classifier being input into other classifiers to see if they also induce changes in the outputs of these classifiers. The relationship between these responses can be quantified and used to identify specific instances where a spurious correlation exists. This is validated by observing intuitive trends in face-attribute and waterbird classifiers, as well as by fabricating spurious correlations and detecting their presence, both visually and quantitatively. Furthermore, utilizing the CF alignment method, we demonstrate that we can evaluate robust optimization methods (GroupDRO, JTT, and FLAC) by detecting a reduction in spurious correlations.
comment: Accepted to Transactions on Machine Learning Research (TMLR), Code: https://github.com/ieee8023/latentshift
♻ ☆ Integrated Push-and-Pull Update Model for Goal-Oriented Effective Communication
This paper studies decision-making for goal-oriented effective communication. We consider an end-to-end status update system where a sensing agent (SA) observes a source, generates and transmits updates to an actuation agent (AA), while the AA takes actions to accomplish a goal at the endpoint. We integrate the push- and pull-based update communication models to obtain a push-and-pull model, which allows the transmission controller at the SA to decide to push an update to the AA and the query controller at the AA to pull updates by raising queries at specific time instances. To gauge effectiveness, we utilize a grade of effectiveness (GoE) metric incorporating updates' freshness, usefulness, and timeliness of actions as qualitative attributes. We then derive effect-aware policies to maximize the expected discounted sum of updates' effectiveness subject to induced costs. The effect-aware policy at the SA considers the potential effectiveness of communicated updates at the endpoint, while at the AA, it accounts for the probabilistic evolution of the source and importance of generated updates. Our results show the proposed push-and-pull model outperforms models solely based on push- or pull-based updates both in terms of efficiency and effectiveness. Additionally, using effect-aware policies at both agents enhances effectiveness compared to periodic and/or probabilistic effect-agnostic policies at either or both agents.
comment: Submitted for possible publication
♻ ☆ Taming the Long Tail in Human Mobility Prediction NeurIPS 2024
With the popularity of location-based services, human mobility prediction plays a key role in enhancing personalized navigation, optimizing recommendation systems, and facilitating urban mobility and planning. This involves predicting a user's next POI (point-of-interest) visit using their past visit history. However, the uneven distribution of visitations over time and space, namely the long-tail problem in spatial distribution, makes it difficult for AI models to predict those POIs that are less visited by humans. In light of this issue, we propose the Long-Tail Adjusted Next POI Prediction (LoTNext) framework for mobility prediction, combining a Long-Tailed Graph Adjustment module to reduce the impact of the long-tailed nodes in the user-POI interaction graph and a novel Long-Tailed Loss Adjustment module to adjust loss by logit score and sample weight adjustment strategy. Also, we employ the auxiliary prediction task to enhance generalization and accuracy. Our experiments with two real-world trajectory datasets demonstrate that LoTNext significantly surpasses existing state-of-the-art works.
comment: Accepted by NeurIPS 2024
♻ ☆ The Surprising Ineffectiveness of Pre-Trained Visual Representations for Model-Based Reinforcement Learning NeurIPS 2024
Visual Reinforcement Learning (RL) methods often require extensive amounts of data. As opposed to model-free RL, model-based RL (MBRL) offers a potential solution with efficient data utilization through planning. Additionally, RL lacks generalization capabilities for real-world tasks. Prior work has shown that incorporating pre-trained visual representations (PVRs) enhances sample efficiency and generalization. While PVRs have been extensively studied in the context of model-free RL, their potential in MBRL remains largely unexplored. In this paper, we benchmark a set of PVRs on challenging control tasks in a model-based RL setting. We investigate the data efficiency, generalization capabilities, and the impact of different properties of PVRs on the performance of model-based agents. Our results, perhaps surprisingly, reveal that for MBRL current PVRs are not more sample efficient than learning representations from scratch, and that they do not generalize better to out-of-distribution (OOD) settings. To explain this, we analyze the quality of the trained dynamics model. Furthermore, we show that data diversity and network architecture are the most important contributors to OOD generalization performance.
comment: Published at the 38th Conference on Neural Information Processing Systems (NeurIPS 2024). Project page: https://schneimo.com/pvr4mbrl/
♻ ☆ Learning Optimal Tax Design in Nonatomic Congestion Games NeurIPS
In multiplayer games, self-interested behavior among the players can harm the social welfare. Tax mechanisms are a common method to alleviate this issue and induce socially optimal behavior. In this work, we take the initial step of learning the optimal tax that can maximize social welfare with limited feedback in congestion games. We propose a new type of feedback named \emph{equilibrium feedback}, where the tax designer can only observe the Nash equilibrium after deploying a tax plan. Existing algorithms are not applicable due to the exponentially large tax function space, nonexistence of the gradient, and nonconvexity of the objective. To tackle these challenges, we design a computationally efficient algorithm that leverages several novel components: (1) a piece-wise linear tax to approximate the optimal tax; (2) extra linear terms to guarantee a strongly convex potential function; (3) an efficient subroutine to find the exploratory tax that can provide critical information about the game. The algorithm can find an $\epsilon$-optimal tax with $O(\beta F^2/\epsilon)$ sample complexity, where $\beta$ is the smoothness of the cost function and $F$ is the number of facilities.
comment: 23 pages. Accepted by Conference on Neural Information Processing Systems (NeurIPS) 2024
♻ ☆ Evaluation of Artificial Intelligence Methods for Lead Time Prediction in Non-Cycled Areas of Automotive Production
The present study examines the effectiveness of applying Artificial Intelligence methods in an automotive production environment to predict unknown lead times in a non-cycle-controlled production area. Data structures are analyzed to identify contextual features and then preprocessed using one-hot encoding. Methods selection focuses on supervised machine learning techniques. In supervised learning methods, regression and classification methods are evaluated. Continuous regression based on target size distribution is not feasible. Classification methods analysis shows that Ensemble Learning and Support Vector Machines are the most suitable. Preliminary study results indicate that gradient boosting algorithms LightGBM, XGBoost, and CatBoost yield the best results. After further testing and extensive hyperparameter optimization, the final method choice is the LightGBM algorithm. Depending on feature availability and prediction interval granularity, relative prediction accuracies of up to 90% can be achieved. Further tests highlight the importance of periodic retraining of AI models to accurately represent complex production processes using the database. The research demonstrates that AI methods can be effectively applied to highly variable production data, adding business value by providing an additional metric for various control tasks while outperforming current non AI-based systems.
♻ ☆ Constrained Latent Action Policies for Model-Based Offline Reinforcement Learning NeurIPS 2024
In offline reinforcement learning, a policy is learned using a static dataset in the absence of costly feedback from the environment. In contrast to the online setting, only using static datasets poses additional challenges, such as policies generating out-of-distribution samples. Model-based offline reinforcement learning methods try to overcome these by learning a model of the underlying dynamics of the environment and using it to guide policy search. It is beneficial but, with limited datasets, errors in the model and the issue of value overestimation among out-of-distribution states can worsen performance. Current model-based methods apply some notion of conservatism to the Bellman update, often implemented using uncertainty estimation derived from model ensembles. In this paper, we propose Constrained Latent Action Policies (C-LAP) which learns a generative model of the joint distribution of observations and actions. We cast policy learning as a constrained objective to always stay within the support of the latent action distribution, and use the generative capabilities of the model to impose an implicit constraint on the generated actions. Thereby eliminating the need to use additional uncertainty penalties on the Bellman update and significantly decreasing the number of gradient steps required to learn a policy. We empirically evaluate C-LAP on the D4RL and V-D4RL benchmark, and show that C-LAP is competitive to state-of-the-art methods, especially outperforming on datasets with visual observations.
comment: 38th Conference on Neural Information Processing Systems (NeurIPS 2024)
♻ ☆ Mind the Error! Detection and Localization of Instruction Errors in Vision-and-Language Navigation IROS'24
Vision-and-Language Navigation in Continuous Environments (VLN-CE) is one of the most intuitive yet challenging embodied AI tasks. Agents are tasked to navigate towards a target goal by executing a set of low-level actions, following a series of natural language instructions. All VLN-CE methods in the literature assume that language instructions are exact. However, in practice, instructions given by humans can contain errors when describing a spatial environment due to inaccurate memory or confusion. Current VLN-CE benchmarks do not address this scenario, making the state-of-the-art methods in VLN-CE fragile in the presence of erroneous instructions from human users. For the first time, we propose a novel benchmark dataset that introduces various types of instruction errors considering potential human causes. This benchmark provides valuable insight into the robustness of VLN systems in continuous environments. We observe a noticeable performance drop (up to -25%) in Success Rate when evaluating the state-of-the-art VLN-CE methods on our benchmark. Moreover, we formally define the task of Instruction Error Detection and Localization, and establish an evaluation protocol on top of our benchmark dataset. We also propose an effective method, based on a cross-modal transformer architecture, that achieves the best performance in error detection and localization, compared to baselines. Surprisingly, our proposed method has revealed errors in the validation set of the two commonly used datasets for VLN-CE, i.e., R2R-CE and RxR-CE, demonstrating the utility of our technique in other tasks. Code and dataset available at https://intelligolabs.github.io/R2RIE-CE
comment: 3 figures, 8 pages. Accepted at IROS'24
♻ ☆ Towards a performance characteristic curve for model evaluation: an application in information diffusion prediction
The information diffusion prediction on social networks aims to predict future recipients of a message, with practical applications in marketing and social media. While different prediction models all claim to perform well, general frameworks for performance evaluation remain limited. Here, we aim to identify a performance characteristic curve for a model, which captures its performance on tasks of different complexity. We propose a metric based on information entropy to quantify the randomness in diffusion data. We then identify a scaling pattern between the randomness and the prediction accuracy of the model. By properly adjusting the variables, data points by different sequence lengths, system sizes, and randomness can all collapse into a single curve. The curve captures a model's inherent capability of making correct predictions against increased uncertainty, which we regard as the performance characteristic curve of the model. The validity of the curve is tested by three prediction models in the same family, reaching conclusions in line with existing studies. In addition, we apply the curve to successfully assess the performance of eight state-of-the-art models, providing a clear and comprehensive evaluation even for models that are challenging to differentiate with conventional metrics. Our work reveals a pattern underlying the data randomness and prediction accuracy. The performance characteristic curve provides a new way to evaluate models' performance systematically, and sheds light on future studies on other frameworks for model evaluation.
♻ ☆ Unseen Horizons: Unveiling the Real Capability of LLM Code Generation Beyond the Familiar ICSE 2025
Recently, large language models (LLMs) have shown strong potential in code generation tasks. However, there are still gaps before they can be fully applied in actual software development processes. Accurately assessing the code generation capabilities of large language models has become an important basis for evaluating and improving the models. Some existing works have constructed datasets to evaluate the capabilities of these models. However, the current evaluation process may encounter the illusion of "Specialist in Familiarity", primarily due to three gaps: the exposure of target code, case timeliness, and dependency availability. The fundamental reason for these gaps is that the code in current datasets may have been extensively exposed and exercised during the training phase, and due to the continuous training and development of LLM, their timeliness has been severely compromised. The key to solve the problem is to, as much as possible, evaluate the LLMs using code that they have not encountered before. Thus, the fundamental idea in this paper is to draw on the concept of code obfuscation, changing code at different levels while ensuring the functionality and output. To this end, we build a code-obfuscation based benchmark OBFUSEVAL. We first collect 1,354 raw cases from five real-world projects, including function description and code. Then we use three-level strategy (symbol, structure and semantic) to obfuscate descriptions, code and context dependencies. We evaluate four LLMs on OBFU- SEVAL and compared the effectiveness of different obfuscation strategy. We use official test suites of these projects to evaluate the generated code. The results show that after obfuscation, the average decrease ratio of test pass rate can up to 62.5%.
comment: Accepted by the 47th International Conference on Software Engineering (ICSE 2025)
♻ ☆ Maximizing Uncertainty for Federated learning via Bayesian Optimisation-based Model Poisoning
As we transition from Narrow Artificial Intelligence towards Artificial Super Intelligence, users are increasingly concerned about their privacy and the trustworthiness of machine learning (ML) technology. A common denominator for the metrics of trustworthiness is the quantification of uncertainty inherent in DL algorithms, and specifically in the model parameters, input data, and model predictions. One of the common approaches to address privacy-related issues in DL is to adopt distributed learning such as federated learning (FL), where private raw data is not shared among users. Despite the privacy-preserving mechanisms in FL, it still faces challenges in trustworthiness. Specifically, the malicious users, during training, can systematically create malicious model parameters to compromise the models predictive and generative capabilities, resulting in high uncertainty about their reliability. To demonstrate malicious behaviour, we propose a novel model poisoning attack method named Delphi which aims to maximise the uncertainty of the global model output. We achieve this by taking advantage of the relationship between the uncertainty and the model parameters of the first hidden layer of the local model. Delphi employs two types of optimisation , Bayesian Optimisation and Least Squares Trust Region, to search for the optimal poisoned model parameters, named as Delphi-BO and Delphi-LSTR. We quantify the uncertainty using the KL Divergence to minimise the distance of the predictive probability distribution towards an uncertain distribution of model output. Furthermore, we establish a mathematical proof for the attack effectiveness demonstrated in FL. Numerical results demonstrate that Delphi-BO induces a higher amount of uncertainty than Delphi-LSTR highlighting vulnerability of FL systems to model poisoning attacks.
comment: 14 pages
♻ ☆ MambaLRP: Explaining Selective State Space Sequence Models
Recent sequence modeling approaches using selective state space sequence models, referred to as Mamba models, have seen a surge of interest. These models allow efficient processing of long sequences in linear time and are rapidly being adopted in a wide range of applications such as language modeling, demonstrating promising performance. To foster their reliable use in real-world scenarios, it is crucial to augment their transparency. Our work bridges this critical gap by bringing explainability, particularly Layer-wise Relevance Propagation (LRP), to the Mamba architecture. Guided by the axiom of relevance conservation, we identify specific components in the Mamba architecture, which cause unfaithful explanations. To remedy this issue, we propose MambaLRP, a novel algorithm within the LRP framework, which ensures a more stable and reliable relevance propagation through these components. Our proposed method is theoretically sound and excels in achieving state-of-the-art explanation performance across a diverse range of models and datasets. Moreover, MambaLRP facilitates a deeper inspection of Mamba architectures, uncovering various biases and evaluating their significance. It also enables the analysis of previous speculations regarding the long-range capabilities of Mamba models.
♻ ☆ Sparse Low-Ranked Self-Attention Transformer for Remaining Useful Lifetime Prediction of Optical Fiber Amplifiers
Optical fiber amplifiers are key elements in present optical networks. Failures of these components result in high financial loss of income of the network operator as the communication traffic over an affected link is interrupted. Applying Remaining useful lifetime (RUL) prediction in the context of Predictive Maintenance (PdM) to optical fiber amplifiers to predict upcoming system failures at an early stage, so that network outages can be minimized through planning of targeted maintenance actions, ensures reliability and safety. Optical fiber amplifier are complex systems, that work under various operating conditions, which makes correct forecasting a difficult task. Increased monitoring capabilities of systems results in datasets that facilitate the application of data-driven RUL prediction methods. Deep learning models in particular have shown good performance, but generalization based on comparatively small datasets for RUL prediction is difficult. In this paper, we propose Sparse Low-ranked self-Attention Transformer (SLAT) as a novel RUL prediction method. SLAT is based on an encoder-decoder architecture, wherein two parallel working encoders extract features for sensors and time steps. By utilizing the self-attention mechanism, long-term dependencies can be learned from long sequences. The implementation of sparsity in the attention matrix and a low-rank parametrization reduce overfitting and increase generalization. Experimental application to optical fiber amplifiers exemplified on EDFA, as well as a reference dataset from turbofan engines, shows that SLAT outperforms the state-of-the-art methods.
comment: 9 pages, 7 figures
♻ ☆ FADE: Towards Fairness-aware Augmentation for Domain Generalization via Classifier-Guided Score-based Diffusion Models
Fairness-aware domain generalization (FairDG) has emerged as a critical challenge for deploying trustworthy AI systems, particularly in scenarios involving distribution shifts. Traditional methods for addressing fairness have failed in domain generalization due to their lack of consideration for distribution shifts. Although disentanglement has been used to tackle FairDG, it is limited by its strong assumptions. To overcome these limitations, we propose Fairness-aware Classifier-Guided Score-based Diffusion Models (FADE) as a novel approach to effectively address the FairDG issue. Specifically, we first pre-train a score-based diffusion model (SDM) and two classifiers to equip the model with strong generalization capabilities across different domains. Then, we guide the SDM using these pre-trained classifiers to effectively eliminate sensitive information from the generated data. Finally, the generated fair data is used to train downstream classifiers, ensuring robust performance under new data distributions. Extensive experiments on three real-world datasets demonstrate that FADE not only enhances fairness but also improves accuracy in the presence of distribution shifts. Additionally, FADE outperforms existing methods in achieving the best accuracy-fairness trade-offs.
♻ ☆ Let Network Decide What to Learn: Symbolic Music Understanding Model Based on Large-scale Adversarial Pre-training
As a crucial aspect of Music Information Retrieval (MIR), Symbolic Music Understanding (SMU) has garnered significant attention for its potential to assist both musicians and enthusiasts in learning and creating music. Recently, pre-trained language models have been widely adopted in SMU due to the substantial similarities between symbolic music and natural language, as well as the ability of these models to leverage limited music data effectively. However, some studies have shown the common pre-trained methods like Mask Language Model (MLM) may introduce bias issues like racism discrimination in Natural Language Process (NLP) and affects the performance of downstream tasks, which also happens in SMU. This bias often arises when masked tokens cannot be inferred from their context, forcing the model to overfit the training set instead of generalizing. To address this challenge, we propose Adversarial-MidiBERT for SMU, which adaptively determines what to mask during MLM via a masker network, rather than employing random masking. By avoiding the masking of tokens that are difficult to infer from context, our model is better equipped to capture contextual structures and relationships, rather than merely conforming to the training data distribution. We evaluate our method across four SMU tasks, and our approach demonstrates excellent performance in all cases. The code for our model is publicly available at https://github.com/RS2002/Adversarial-MidiBERT.
♻ ☆ Trustworthy, Responsible, and Safe AI: A Comprehensive Architectural Framework for AI Safety with Challenges and Mitigations
AI Safety is an emerging area of critical importance to the safe adoption and deployment of AI systems. With the rapid proliferation of AI and especially with the recent advancement of Generative AI (or GAI), the technology ecosystem behind the design, development, adoption, and deployment of AI systems has drastically changed, broadening the scope of AI Safety to address impacts on public safety and national security. In this paper, we propose a novel architectural framework for understanding and analyzing AI Safety; defining its characteristics from three perspectives: Trustworthy AI, Responsible AI, and Safe AI. We provide an extensive review of current research and advancements in AI safety from these perspectives, highlighting their key challenges and mitigation approaches. Through examples from state-of-the-art technologies, particularly Large Language Models (LLMs), we present innovative mechanism, methodologies, and techniques for designing and testing AI safety. Our goal is to promote advancement in AI safety research, and ultimately enhance people's trust in digital transformation.
♻ ☆ Diffusion-based Unsupervised Audio-visual Speech Enhancement
This paper proposes a new unsupervised audio-visual speech enhancement (AVSE) approach that combines a diffusion-based audio-visual speech generative model with a non-negative matrix factorization (NMF) noise model. First, the diffusion model is pre-trained on clean speech conditioned on corresponding video data to simulate the speech generative distribution. This pre-trained model is then paired with the NMF-based noise model to estimate clean speech iteratively. Specifically, a diffusion-based posterior sampling approach is implemented within the reverse diffusion process, where after each iteration, a speech estimate is obtained and used to update the noise parameters. Experimental results confirm that the proposed AVSE approach not only outperforms its audio-only counterpart but also generalizes better than a recent supervised-generative AVSE method. Additionally, the new inference algorithm offers a better balance between inference speed and performance compared to the previous diffusion-based method. Code and demo available at: https://jeaneudesayilo.github.io/fast_UdiffSE
♻ ☆ Improving Pain Classification using Spatio-Temporal Deep Learning Approaches with Facial Expressions
Pain management and severity detection are crucial for effective treatment, yet traditional self-reporting methods are subjective and may be unsuitable for non-verbal individuals (people with limited speaking skills). To address this limitation, we explore automated pain detection using facial expressions. Our study leverages deep learning techniques to improve pain assessment by analyzing facial images from the Pain Emotion Faces Database (PEMF). We propose two novel approaches1: (1) a hybrid ConvNeXt model combined with Long Short-Term Memory (LSTM) blocks to analyze video frames and predict pain presence, and (2) a Spatio-Temporal Graph Convolution Network (STGCN) integrated with LSTM to process landmarks from facial images for pain detection. Our work represents the first use of the PEMF dataset for binary pain classification and demonstrates the effectiveness of these models through extensive experimentation. The results highlight the potential of combining spatial and temporal features for enhanced pain detection, offering a promising advancement in objective pain assessment methodologies.
comment: 8 pages, 3 figures, 3 tables. Accepted and presented at the 18th International Conference on Machine Vision (ICMV 2024), Edinburgh, UK
♻ ☆ SupplyGraph: A Benchmark Dataset for Supply Chain Planning using Graph Neural Networks AAAI 2024
Graph Neural Networks (GNNs) have gained traction across different domains such as transportation, bio-informatics, language processing, and computer vision. However, there is a noticeable absence of research on applying GNNs to supply chain networks. Supply chain networks are inherently graph-like in structure, making them prime candidates for applying GNN methodologies. This opens up a world of possibilities for optimizing, predicting, and solving even the most complex supply chain problems. A major setback in this approach lies in the absence of real-world benchmark datasets to facilitate the research and resolution of supply chain problems using GNNs. To address the issue, we present a real-world benchmark dataset for temporal tasks, obtained from one of the leading FMCG companies in Bangladesh, focusing on supply chain planning for production purposes. The dataset includes temporal data as node features to enable sales predictions, production planning, and the identification of factory issues. By utilizing this dataset, researchers can employ GNNs to address numerous supply chain problems, thereby advancing the field of supply chain analytics and planning. Source: https://github.com/CIOL-SUST/SupplyGraph
comment: Accepted to 4th workshop on Graphs and more Complex structures for Learning and Reasoning, colocated with AAAI 2024
♻ ☆ Get Rid of Isolation: A Continuous Multi-task Spatio-Temporal Learning Framework NeurIPS 2024
Spatiotemporal learning has become a pivotal technique to enable urban intelligence. Traditional spatiotemporal models mostly focus on a specific task by assuming a same distribution between training and testing sets. However, given that urban systems are usually dynamic, multi-sourced with imbalanced data distributions, current specific task-specific models fail to generalize to new urban conditions and adapt to new domains without explicitly modeling interdependencies across various dimensions and types of urban data. To this end, we argue that there is an essential to propose a Continuous Multi-task Spatio-Temporal learning framework (CMuST) to empower collective urban intelligence, which reforms the urban spatiotemporal learning from single-domain to cooperatively multi-dimensional and multi-task learning. Specifically, CMuST proposes a new multi-dimensional spatiotemporal interaction network (MSTI) to allow cross-interactions between context and main observations as well as self-interactions within spatial and temporal aspects to be exposed, which is also the core for capturing task-level commonality and personalization. To ensure continuous task learning, a novel Rolling Adaptation training scheme (RoAda) is devised, which not only preserves task uniqueness by constructing data summarization-driven task prompts, but also harnesses correlated patterns among tasks by iterative model behavior modeling. We further establish a benchmark of three cities for multi-task spatiotemporal learning, and empirically demonstrate the superiority of CMuST via extensive evaluations on these datasets. The impressive improvements on both few-shot streaming data and new domain tasks against existing SOAT methods are achieved. Code is available at https://github.com/DILab-USTCSZ/CMuST.
comment: Accepted by NeurIPS 2024
♻ ☆ Toward Automated Simulation Research Workflow through LLM Prompt Engineering Design
The advent of Large Language Models (LLMs) has created new opportunities for the automation of scientific research spanning both experimental processes and computational simulations. This study explores the feasibility of constructing an autonomous simulation agent (ASA) powered by LLMs through prompt engineering and automated program design to automate the entire simulation research process according to a human-provided research plan. This process includes experimental design, remote upload and simulation execution, data analysis, and report compilation. Using a well-studied simulation problem of polymer chain conformations as a test case, we assessed the long-task completion and reliability of ASAs powered by different LLMs, including GPT-4o, Claude-3.5, etc. Our findings revealed that ASA-GPT-4o achieved near-flawless execution on designated research missions, underscoring the potential of methods like ASA to achieve automation in simulation research processes to enhance research efficiency. The outlined automation can be iteratively performed for up to 20 cycles without human intervention, illustrating the potential of ASA for long-task workflow automation. Additionally, we discussed the intrinsic traits of ASA in managing extensive tasks, focusing on self-validation mechanisms, and the balance between local attention and global oversight.
comment: The source code and example results of ASA can be found at https://github.com/zokaraa/autonomous_simulation_agent
♻ ☆ Fully Distributed, Flexible Compositional Visual Representations via Soft Tensor Products
Since the inception of the classicalist vs. connectionist debate, it has been argued that the ability to systematically combine symbol-like entities into compositional representations is crucial for human intelligence. In connectionist systems, the field of disentanglement has gained prominence for its ability to produce explicitly compositional representations; however, it relies on a fundamentally symbolic, concatenative representation of compositional structure that clashes with the continuous, distributed foundations of deep learning. To resolve this tension, we extend Smolensky's Tensor Product Representation (TPR) and introduce Soft TPR, a representational form that encodes compositional structure in an inherently distributed, flexible manner, along with Soft TPR Autoencoder, a theoretically-principled architecture designed specifically to learn Soft TPRs. Comprehensive evaluations in the visual representation learning domain demonstrate that the Soft TPR framework consistently outperforms conventional disentanglement alternatives -- achieving state-of-the-art disentanglement, boosting representation learner convergence, and delivering superior sample efficiency and low-sample regime performance in downstream tasks. These findings highlight the promise of a distributed and flexible approach to representing compositional structure by potentially enhancing alignment with the core principles of deep learning over the conventional symbolic approach.
comment: Accepted to Neurips 2024. 10 pages + supplementary
♻ ☆ SelectIT: Selective Instruction Tuning for LLMs via Uncertainty-Aware Self-Reflection NeurIPS 2024
Instruction tuning (IT) is crucial to tailoring large language models (LLMs) towards human-centric interactions. Recent advancements have shown that the careful selection of a small, high-quality subset of IT data can significantly enhance the performance of LLMs. Despite this, common approaches often rely on additional models or data, which increases costs and limits widespread adoption. In this work, we propose a novel approach, termed SelectIT, that capitalizes on the foundational capabilities of the LLM itself. Specifically, we exploit the intrinsic uncertainty present in LLMs to more effectively select high-quality IT data, without the need for extra resources. Furthermore, we introduce a curated IT dataset, the Selective Alpaca, created by applying SelectIT to the Alpaca-GPT4 dataset. Empirical results demonstrate that IT using Selective Alpaca leads to substantial model ability enhancement. The robustness of SelectIT has also been corroborated in various foundation models and domain-specific tasks. Our findings suggest that longer and more computationally intensive IT data may serve as superior sources of IT, offering valuable insights for future research in this area. Data, code, and scripts are freely available at https://github.com/Blue-Raincoat/SelectIT.
comment: Accepted to NeurIPS 2024
♻ ☆ Making AI Less "Thirsty": Uncovering and Addressing the Secret Water Footprint of AI Models
The growing carbon footprint of artificial intelligence (AI) has been undergoing public scrutiny. Nonetheless, the equally important water (withdrawal and consumption) footprint of AI has largely remained under the radar. For example, training the GPT-3 language model in Microsoft's state-of-the-art U.S. data centers can directly evaporate 700,000 liters of clean freshwater, but such information has been kept a secret. More critically, the global AI demand is projected to account for 4.2-6.6 billion cubic meters of water withdrawal in 2027, which is more than the total annual water withdrawal of 4-6 Denmark or half of the United Kingdom. This is concerning, as freshwater scarcity has become one of the most pressing challenges. To respond to the global water challenges, AI can, and also must, take social responsibility and lead by example by addressing its own water footprint. In this paper, we provide a principled methodology to estimate the water footprint of AI, and also discuss the unique spatial-temporal diversities of AI's runtime water efficiency. Finally, we highlight the necessity of holistically addressing water footprint along with carbon footprint to enable truly sustainable AI.
comment: Accepted by Communications of the ACM. Source codes available at: https://github.com/Ren-Research/Making-AI-Less-Thirsty
♻ ☆ Mitigating Knowledge Conflicts in Language Model-Driven Question Answering
In the context of knowledge-driven seq-to-seq generation tasks, such as document-based question answering and document summarization systems, two fundamental knowledge sources play crucial roles: the inherent knowledge embedded within model parameters and the external knowledge obtained through context. Recent studies revealed a significant challenge: when there exists a misalignment between the model's inherent knowledge and the ground truth answers in training data, the system may exhibit problematic behaviors during inference, such as ignoring input context, or generating unfaithful content. Our investigation proposes a strategy to minimize hallucination by building explicit connection between source inputs and generated outputs. We specifically target a common hallucination pattern in question answering, examining how the correspondence between entities and their contexts during model training influences the system's performance at inference time.
comment: revised version, more figures
♻ ☆ OminiControl: Minimal and Universal Control for Diffusion Transformer
In this paper, we introduce OminiControl, a highly versatile and parameter-efficient framework that integrates image conditions into pre-trained Diffusion Transformer (DiT) models. At its core, OminiControl leverages a parameter reuse mechanism, enabling the DiT to encode image conditions using itself as a powerful backbone and process them with its flexible multi-modal attention processors. Unlike existing methods, which rely heavily on additional encoder modules with complex architectures, OminiControl (1) effectively and efficiently incorporates injected image conditions with only ~0.1% additional parameters, and (2) addresses a wide range of image conditioning tasks in a unified manner, including subject-driven generation and spatially-aligned conditions such as edges, depth, and more. Remarkably, these capabilities are achieved by training on images generated by the DiT itself, which is particularly beneficial for subject-driven generation. Extensive evaluations demonstrate that OminiControl outperforms existing UNet-based and DiT-adapted models in both subject-driven and spatially-aligned conditional generation. Additionally, we release our training dataset, Subjects200K, a diverse collection of over 200,000 identity-consistent images, along with an efficient data synthesis pipeline to advance research in subject-consistent generation.
♻ ☆ CrossFi: A Cross Domain Wi-Fi Sensing Framework Based on Siamese Network
In recent years, Wi-Fi sensing has garnered significant attention due to its numerous benefits, such as privacy protection, low cost, and penetration ability. Extensive research has been conducted in this field, focusing on areas such as gesture recognition, people identification, and fall detection. However, many data-driven methods encounter challenges related to domain shift, where the model fails to perform well in environments different from the training data. One major factor contributing to this issue is the limited availability of Wi-Fi sensing datasets, which makes models learn excessive irrelevant information and over-fit to the training set. Unfortunately, collecting large-scale Wi-Fi sensing datasets across diverse scenarios is a challenging task. To address this problem, we propose CrossFi, a siamese network-based approach that excels in both in-domain scenario and cross-domain scenario, including few-shot, zero-shot scenarios, and even works in few-shot new-class scenario where testing set contains new categories. The core component of CrossFi is a sample-similarity calculation network called CSi-Net, which improves the structure of the siamese network by using an attention mechanism to capture similarity information, instead of simply calculating the distance or cosine similarity. Based on it, we develop an extra Weight-Net that can generate a template for each class, so that our CrossFi can work in different scenarios. Experimental results demonstrate that our CrossFi achieves state-of-the-art performance across various scenarios. In gesture recognition task, our CrossFi achieves an accuracy of 98.17% in in-domain scenario, 91.72% in one-shot cross-domain scenario, 64.81% in zero-shot cross-domain scenario, and 84.75% in one-shot new-class scenario. The code for our model is publicly available at https://github.com/RS2002/CrossFi.
♻ ☆ The Silent Majority: Demystifying Memorization Effect in the Presence of Spurious Correlations
Machine learning models often rely on simple spurious features -- patterns in training data that correlate with targets but are not causally related to them, like image backgrounds in foreground classification. This reliance typically leads to imbalanced test performance across minority and majority groups. In this work, we take a closer look at the fundamental cause of such imbalanced performance through the lens of memorization, which refers to the ability to predict accurately on \textit{atypical} examples (minority groups) in the training set but failing in achieving the same accuracy in the testing set. This paper systematically shows the ubiquitous existence of spurious features in a small set of neurons within the network, providing the first-ever evidence that memorization may contribute to imbalanced group performance. Through three experimental sources of converging empirical evidence, we find the property of a small subset of neurons or channels in memorizing minority group information. Inspired by these findings, we articulate the hypothesis: the imbalanced group performance is a byproduct of ``noisy'' spurious memorization confined to a small set of neurons. To further substantiate this hypothesis, we show that eliminating these unnecessary spurious memorization patterns via a novel framework during training can significantly affect the model performance on minority groups. Our experimental results across various architectures and benchmarks offer new insights on how neural networks encode core and spurious knowledge, laying the groundwork for future research in demystifying robustness to spurious correlation.
♻ ☆ Noise-powered Multi-modal Knowledge Graph Representation Framework COLING 2025
The rise of Multi-modal Pre-training highlights the necessity for a unified Multi-Modal Knowledge Graph (MMKG) representation learning framework. Such a framework is essential for embedding structured knowledge into multi-modal Large Language Models effectively, alleviating issues like knowledge misconceptions and multi-modal hallucinations. In this work, we explore the efficacy of models in accurately embedding entities within MMKGs through two pivotal tasks: Multi-modal Knowledge Graph Completion (MKGC) and Multi-modal Entity Alignment (MMEA). Building on this foundation, we propose a novel SNAG method that utilizes a Transformer-based architecture equipped with modality-level noise masking to robustly integrate multi-modal entity features in KGs. By incorporating specific training objectives for both MKGC and MMEA, our approach achieves SOTA performance across a total of ten datasets, demonstrating its versatility. Moreover, SNAG can not only function as a standalone model but also enhance other existing methods, providing stable performance improvements. Code and data are available at https://github.com/zjukg/SNAG.
comment: COLING 2025 Accepted, Repo is available at https://github.com/zjukg/SNAG
♻ ☆ Machine unlearning through fine-grained model parameters perturbation
Machine unlearning techniques, which involve retracting data records and reducing influence of said data on trained models, help with the user privacy protection objective but incur significant computational costs. Weight perturbation-based unlearning is a general approach, but it typically involves globally modifying the parameters. We propose fine-grained Top-K and Random-k parameters perturbed inexact machine unlearning strategies that address the privacy needs while keeping the computational costs tractable. In order to demonstrate the efficacy of our strategies we also tackle the challenge of evaluating the effectiveness of machine unlearning by considering the model's generalization performance across both unlearning and remaining data. To better assess the unlearning effect and model generalization, we propose novel metrics, namely, the forgetting rate and memory retention rate. However, for inexact machine unlearning, current metrics are inadequate in quantifying the degree of forgetting that occurs after unlearning strategies are applied. To address this, we introduce SPD-GAN, which subtly perturbs the distribution of data targeted for unlearning. Then, we evaluate the degree of unlearning by measuring the performance difference of the models on the perturbed unlearning data before and after the unlearning process. By implementing these innovative techniques and metrics, we achieve computationally efficacious privacy protection in machine learning applications without significant sacrifice of model performance. Furthermore, this approach provides a novel method for evaluating the degree of unlearning.
♻ ☆ STORM: A Spatio-Temporal Factor Model Based on Dual Vector Quantized Variational Autoencoders for Financial Trading
In financial trading, factor models are widely used to price assets and capture excess returns from mispricing. Recently, we have witnessed the rise of variational autoencoder-based latent factor models, which learn latent factors self-adaptively. While these models focus on modeling overall market conditions, they often fail to effectively capture the temporal patterns of individual stocks. Additionally, representing multiple factors as single values simplifies the model but limits its ability to capture complex relationships and dependencies. As a result, the learned factors are of low quality and lack diversity, reducing their effectiveness and robustness across different trading periods. To address these issues, we propose a Spatio-Temporal factOR Model based on dual vector quantized variational autoencoders, named STORM, which extracts features of stocks from temporal and spatial perspectives, then fuses and aligns these features at the fine-grained and semantic level, and represents the factors as multi-dimensional embeddings. The discrete codebooks cluster similar factor embeddings, ensuring orthogonality and diversity, which helps distinguish between different factors and enables factor selection in financial trading. To show the performance of the proposed factor model, we apply it to two downstream experiments: portfolio management on two stock datasets and individual trading tasks on six specific stocks. The extensive experiments demonstrate STORM's flexibility in adapting to downstream tasks and superior performance over baseline models.
♻ ☆ Do Large Language Models Mirror Cognitive Language Processing?
Large Language Models (LLMs) have demonstrated remarkable abilities in text comprehension and logical reasoning, indicating that the text representations learned by LLMs can facilitate their language processing capabilities. In neuroscience, brain cognitive processing signals are typically utilized to study human language processing. Therefore, it is natural to ask how well the text embeddings from LLMs align with the brain cognitive processing signals, and how training strategies affect the LLM-brain alignment? In this paper, we employ Representational Similarity Analysis (RSA) to measure the alignment between 23 mainstream LLMs and fMRI signals of the brain to evaluate how effectively LLMs simulate cognitive language processing. We empirically investigate the impact of various factors (e.g., pre-training data size, model scaling, alignment training, and prompts) on such LLM-brain alignment. Experimental results indicate that pre-training data size and model scaling are positively correlated with LLM-brain similarity, and alignment training can significantly improve LLM-brain similarity. Explicit prompts contribute to the consistency of LLMs with brain cognitive language processing, while nonsensical noisy prompts may attenuate such alignment. Additionally, the performance of a wide range of LLM evaluations (e.g., MMLU, Chatbot Arena) is highly correlated with the LLM-brain similarity.
♻ ☆ EdgeSight: Enabling Modeless and Cost-Efficient Inference at the Edge
Traditional ML inference is evolving toward modeless inference, which abstracts the complexity of model selection from users, allowing the system to automatically choose the most appropriate model for each request based on accuracy and resource requirements. While prior studies have focused on modeless inference within data centers, this paper tackles the pressing need for cost-efficient modeless inference at the edge -- particularly within its unique constraints of limited device memory, volatile network conditions, and restricted power consumption. To overcome these challenges, we propose EdgeSight, a system that provides cost-efficient EdgeSight serving for diverse DNNs at the edge. EdgeSight employs an edge-data center (edge-DC) architecture, utilizing confidence scaling to reduce the number of model options while meeting diverse accuracy requirements. Additionally, it supports lossy inference in volatile network environments. Our experimental results show that EdgeSight outperforms existing systems by up to 1.6x in P99 latency for modeless services. Furthermore, our FPGA prototype demonstrates similar performance at certain accuracy levels, with a power consumption reduction of up to 3.34x.
comment: 12 pages
♻ ☆ Natural Language Outlines for Code: Literate Programming in the LLM Era
We propose using natural language outlines as a novel modality and interaction surface for providing AI assistance to developers throughout the software development process. An NL outline for a code function comprises multiple statements written in concise prose, which partition the code and summarize its main ideas in the style of literate programming. Crucially, we find that modern LLMs can generate accurate and high-quality NL outlines in practice. Moreover, NL outlines enable a bidirectional sync between code and NL, allowing changes in one to be automatically reflected in the other. We discuss many use cases for NL outlines: they can accelerate understanding and navigation of code and diffs, simplify code maintenance, augment code search, steer code generation, and more. We then propose and compare multiple LLM prompting techniques for generating outlines and ask professional developers to judge outline quality. Finally, we present two case studies applying NL outlines toward code review and malware detection.
♻ ☆ Continual Diffuser (CoD): Mastering Continual Offline Reinforcement Learning with Experience Rehearsal
Artificial neural networks, especially recent diffusion-based models, have shown remarkable superiority in gaming, control, and QA systems, where the training tasks' datasets are usually static. However, in real-world applications, such as robotic control of reinforcement learning (RL), the tasks are changing, and new tasks arise in a sequential order. This situation poses the new challenge of plasticity-stability trade-off for training an agent who can adapt to task changes and retain acquired knowledge. In view of this, we propose a rehearsal-based continual diffusion model, called Continual Diffuser (CoD), to endow the diffuser with the capabilities of quick adaptation (plasticity) and lasting retention (stability). Specifically, we first construct an offline benchmark that contains 90 tasks from multiple domains. Then, we train the CoD on each task with sequential modeling and conditional generation for making decisions. Next, we preserve a small portion of previous datasets as the rehearsal buffer and replay it to retain the acquired knowledge. Extensive experiments on a series of tasks show CoD can achieve a promising plasticity-stability trade-off and outperform existing diffusion-based methods and other representative baselines on most tasks.
comment: This work has been submitted to the IEEE for possible publication
♻ ☆ A Multi-Modal AI Copilot for Single-Cell Analysis with Instruction Following
Large language models excel at interpreting complex natural language instructions, enabling them to perform a wide range of tasks. In the life sciences, single-cell RNA sequencing (scRNA-seq) data serves as the "language of cellular biology", capturing intricate gene expression patterns at the single-cell level. However, interacting with this "language" through conventional tools is often inefficient and unintuitive, posing challenges for researchers. To address these limitations, we present InstructCell, a multi-modal AI copilot that leverages natural language as a medium for more direct and flexible single-cell analysis. We construct a comprehensive multi-modal instruction dataset that pairs text-based instructions with scRNA-seq profiles from diverse tissues and species. Building on this, we develop a multi-modal cell language architecture capable of simultaneously interpreting and processing both modalities. InstructCell empowers researchers to accomplish critical tasks-such as cell type annotation, conditional pseudo-cell generation, and drug sensitivity prediction-using straightforward natural language commands. Extensive evaluations demonstrate that InstructCell consistently meets or exceeds the performance of existing single-cell foundation models, while adapting to diverse experimental conditions. More importantly, InstructCell provides an accessible and intuitive tool for exploring complex single-cell data, lowering technical barriers and enabling deeper biological insights.
comment: 37 pages; 13 figures; Code: https://github.com/zjunlp/Instructcell, Models: https://huggingface.co/zjunlp/Instructcell-chat, https://huggingface.co/zjunlp/InstructCell-instruct
♻ ☆ Understanding Emergent Abilities of Language Models from the Loss Perspective NeurIPS 2024
Recent studies have put into question the belief that emergent abilities in language models are exclusive to large models. This skepticism arises from two observations: 1) smaller models can also exhibit high performance on emergent abilities and 2) there is doubt on the discontinuous metrics used to measure these abilities. In this paper, we propose to study emergent abilities in the lens of pre-training loss, instead of model size or training compute. We demonstrate that the Transformer models with the same pre-training loss, but different model and data sizes, generate the same performance on various downstream tasks, with a fixed data corpus, tokenization, and model architecture. We also discover that a model exhibits emergent abilities on certain tasks -- regardless of the continuity of metrics -- when its pre-training loss falls below a specific threshold. Before reaching this threshold, its performance remains at the level of random guessing. This inspires us to redefine emergent abilities as those that manifest in models with lower pre-training losses, highlighting that these abilities cannot be predicted by merely extrapolating the performance trends of models with higher pre-training losses.
comment: 23 pages, 8 figures. Accepted in NeurIPS 2024
♻ ☆ Data-driven inventory management for new products: A warm-start and adjusted Dyna-$Q$ approach
In this paper, we propose a novel reinforcement learning algorithm for inventory management of newly launched products with no or limited historical demand information. The algorithm follows the classic Dyna-$Q$ structure, balancing the model-based and model-free approaches, while accelerating the training process of Dyna-$Q$ and mitigating the model discrepancy generated by the model-based feedback. Warm-start information from the demand data of existing similar products can be incorporated into the algorithm to further stabilize the early-stage training and reduce the variance of the estimated optimal policy. Our approach is validated through a case study of bakery inventory management with real data. The adjusted Dyna-$Q$ shows up to a 23.7% reduction in average daily cost compared with $Q$-learning, and up to a 77.5% reduction in training time within the same horizon compared with classic Dyna-$Q$. By incorporating the warm-start information, it can be found that the adjusted Dyna-$Q$ has the lowest total cost, lowest variance in total cost, and relatively low shortage percentages among all the algorithms under a 30-day testing.
comment: 7 pages, 2 figures
♻ ☆ CSL-L2M: Controllable Song-Level Lyric-to-Melody Generation Based on Conditional Transformer with Fine-Grained Lyric and Musical Controls AAAI-25
Lyric-to-melody generation is a highly challenging task in the field of AI music generation. Due to the difficulty of learning strict yet weak correlations between lyrics and melodies, previous methods have suffered from weak controllability, low-quality and poorly structured generation. To address these challenges, we propose CSL-L2M, a controllable song-level lyric-to-melody generation method based on an in-attention Transformer decoder with fine-grained lyric and musical controls, which is able to generate full-song melodies matched with the given lyrics and user-specified musical attributes. Specifically, we first introduce REMI-Aligned, a novel music representation that incorporates strict syllable- and sentence-level alignments between lyrics and melodies, facilitating precise alignment modeling. Subsequently, sentence-level semantic lyric embeddings independently extracted from a sentence-wise Transformer encoder are combined with word-level part-of-speech embeddings and syllable-level tone embeddings as fine-grained controls to enhance the controllability of lyrics over melody generation. Then we introduce human-labeled musical tags, sentence-level statistical musical attributes, and learned musical features extracted from a pre-trained VQ-VAE as coarse-grained, fine-grained and high-fidelity controls, respectively, to the generation process, thereby enabling user control over melody generation. Finally, an in-attention Transformer decoder technique is leveraged to exert fine-grained control over the full-song melody generation with the aforementioned lyric and musical conditions. Experimental results demonstrate that our proposed CSL-L2M outperforms the state-of-the-art models, generating melodies with higher quality, better controllability and enhanced structure. Demos and source code are available at https://lichaiustc.github.io/CSL-L2M/.
comment: Accepted at AAAI-25
♻ ☆ Unconditional stability of a recurrent neural circuit implementing divisive normalization
Stability in recurrent neural models poses a significant challenge, particularly in developing biologically plausible neurodynamical models that can be seamlessly trained. Traditional cortical circuit models are notoriously difficult to train due to expansive nonlinearities in the dynamical system, leading to an optimization problem with nonlinear stability constraints that are difficult to impose. Conversely, recurrent neural networks (RNNs) excel in tasks involving sequential data but lack biological plausibility and interpretability. In this work, we address these challenges by linking dynamic divisive normalization (DN) to the stability of ORGaNICs, a biologically plausible recurrent cortical circuit model that dynamically achieves DN and that has been shown to simulate a wide range of neurophysiological phenomena. By using the indirect method of Lyapunov, we prove the remarkable property of unconditional local stability for an arbitrary-dimensional ORGaNICs circuit when the recurrent weight matrix is the identity. We thus connect ORGaNICs to a system of coupled damped harmonic oscillators, which enables us to derive the circuit's energy function, providing a normative principle of what the circuit, and individual neurons, aim to accomplish. Further, for a generic recurrent weight matrix, we prove the stability of the 2D model and demonstrate empirically that stability holds in higher dimensions. Finally, we show that ORGaNICs can be trained by backpropagation through time without gradient clipping/scaling, thanks to its intrinsic stability property and adaptive time constants, which address the problems of exploding, vanishing, and oscillating gradients. By evaluating the model's performance on RNN benchmarks, we find that ORGaNICs outperform alternative neurodynamical models on static image classification tasks and perform comparably to LSTMs on sequential tasks.
♻ ☆ AI Consciousness is Inevitable: A Theoretical Computer Science Perspective
We look at consciousness through the lens of Theoretical Computer Science, a branch of mathematics that studies computation under resource limitations. From this perspective, we develop a formal machine model for consciousness. The model is inspired by Alan Turing's simple yet powerful model of computation and Bernard Baars' theater model of consciousness. Though extremely simple, the model (1) aligns at a high level with many of the major scientific theories of human and animal consciousness, (2) provides explanations at a high level for many phenomena associated with consciousness, and (3) gives insight into how a machine can have subjective consciousness. This combination supports our claim that machine consciousness is not only plausible but inevitable.
♻ ☆ Multi-modal and Multi-scale Spatial Environment Understanding for Immersive Visual Text-to-Speech AAAI'2025
Visual Text-to-Speech (VTTS) aims to take the environmental image as the prompt to synthesize the reverberant speech for the spoken content. The challenge of this task lies in understanding the spatial environment from the image. Many attempts have been made to extract global spatial visual information from the RGB space of an spatial image. However, local and depth image information are crucial for understanding the spatial environment, which previous works have ignored. To address the issues, we propose a novel multi-modal and multi-scale spatial environment understanding scheme to achieve immersive VTTS, termed M2SE-VTTS. The multi-modal aims to take both the RGB and Depth spaces of the spatial image to learn more comprehensive spatial information, and the multi-scale seeks to model the local and global spatial knowledge simultaneously. Specifically, we first split the RGB and Depth images into patches and adopt the Gemini-generated environment captions to guide the local spatial understanding. After that, the multi-modal and multi-scale features are integrated by the local-aware global spatial understanding. In this way, M2SE-VTTS effectively models the interactions between local and global spatial contexts in the multi-modal spatial environment. Objective and subjective evaluations suggest that our model outperforms the advanced baselines in environmental speech generation. The code and audio samples are available at: https://github.com/AI-S2-Lab/M2SE-VTTS.
comment: 9 pages,2 figures, Accepted by AAAI'2025
♻ ☆ Compositional Automata Embeddings for Goal-Conditioned Reinforcement Learning
Goal-conditioned reinforcement learning is a powerful way to control an AI agent's behavior at runtime. That said, popular goal representations, e.g., target states or natural language, are either limited to Markovian tasks or rely on ambiguous task semantics. We propose representing temporal goals using compositions of deterministic finite automata (cDFAs) and use cDFAs to guide RL agents. cDFAs balance the need for formal temporal semantics with ease of interpretation: if one can understand a flow chart, one can understand a cDFA. On the other hand, cDFAs form a countably infinite concept class with Boolean semantics, and subtle changes to the automaton can result in very different tasks, making them difficult to condition agent behavior on. To address this, we observe that all paths through a DFA correspond to a series of reach-avoid tasks and propose pre-training graph neural network embeddings on "reach-avoid derived" DFAs. Through empirical evaluation, we demonstrate that the proposed pre-training method enables zero-shot generalization to various cDFA task classes and accelerated policy specialization without the myopic suboptimality of hierarchical methods.
♻ ☆ LoL-PIM: Long-Context LLM Decoding with Scalable DRAM-PIM System
The expansion of large language models (LLMs) with hundreds of billions of parameters presents significant challenges to computational resources, particularly data movement and memory bandwidth. Long-context LLMs, which process sequences of tens of thousands of tokens, further increase the demand on the memory system as the complexity in attention layers and key-value cache sizes is proportional to the context length. Processing-in-Memory (PIM) maximizes memory bandwidth by moving compute to the data and can address the memory bandwidth challenges; however, PIM is not necessarily scalable to accelerate long-context LLM because of limited per-module memory capacity and the inflexibility of fixed-functional unit PIM architecture and static memory management. In this work, we propose LoL-PIM which is a multi-node PIM architecture that accelerates long context LLM through hardware-software co-design. In particular, we propose how pipeline parallelism can be exploited across a multi-PIM module while a direct PIM access (DPA) controller (or DMA for PIM) is proposed that enables dynamic PIM memory management and results in efficient PIM utilization across a diverse range of context length. We developed an MLIR-based compiler for LoL-PIM extending a commercial PIM-based compiler where the software modifications were implemented and evaluated, while the hardware changes were modeled in the simulator. Our evaluations demonstrate that LoL-PIM significantly improves throughput and reduces latency for long-context LLM inference, outperforming both multi-GPU and GPU-PIM systems (up to 8.54x and 16.0x speedup, respectively), thereby enabling more efficient deployment of LLMs in real-world applications.
comment: 15 pages, 12 figures
♻ ☆ Learning Cross-Domain Representations for Transferable Drug Perturbations on Single-Cell Transcriptional Responses AAAI
Phenotypic drug discovery has attracted widespread attention because of its potential to identify bioactive molecules. Transcriptomic profiling provides a comprehensive reflection of phenotypic changes in cellular responses to external perturbations. In this paper, we propose XTransferCDR, a novel generative framework designed for feature decoupling and transferable representation learning across domains. Given a pair of perturbed expression profiles, our approach decouples the perturbation representations from basal states through domain separation encoders and then cross-transfers them in the latent space. The transferred representations are then used to reconstruct the corresponding perturbed expression profiles via a shared decoder. This cross-transfer constraint effectively promotes the learning of transferable drug perturbation representations. We conducted extensive evaluations of our model on multiple datasets, including single-cell transcriptional responses to drugs and single- and combinatorial genetic perturbations. The experimental results show that XTransferCDR achieved better performance than current state-of-the-art methods, showcasing its potential to advance phenotypic drug discovery.
comment: Accepted by The 39th Annual AAAI Conference on Artificial Intelligenc (AAAI 2025)
♻ ☆ Efficient Long Video Tokenization via Coordinate-based Patch Reconstruction
Efficient tokenization of videos remains a challenge in training vision models that can process long videos. One promising direction is to develop a tokenizer that can encode long video clips, as it would enable the tokenizer to leverage the temporal coherence of videos better for tokenization. However, training existing tokenizers on long videos often incurs a huge training cost as they are trained to reconstruct all the frames at once. In this paper, we introduce CoordTok, a video tokenizer that learns a mapping from coordinate-based representations to the corresponding patches of input videos, inspired by recent advances in 3D generative models. In particular, CoordTok encodes a video into factorized triplane representations and reconstructs patches that correspond to randomly sampled $(x,y,t)$ coordinates. This allows for training large tokenizer models directly on long videos without requiring excessive training resources. Our experiments show that CoordTok can drastically reduce the number of tokens for encoding long video clips. For instance, CoordTok can encode a 128-frame video with 128$\times$128 resolution into 1280 tokens, while baselines need 6144 or 8192 tokens to achieve similar reconstruction quality. We further show that this efficient video tokenization enables memory-efficient training of a diffusion transformer that can generate 128 frames at once.
comment: Code is available on the project webpage: https://huiwon-jang.github.io/coordtok/
♻ ☆ Automated Review Generation Method Based on Large Language Models
Literature research, vital for scientific work, faces the challenge of surging information volumes exceeding researchers' processing capabilities. We present an automated review generation method based on large language models (LLMs) to overcome efficiency bottlenecks and reduce cognitive load. Our statistically validated evaluation framework demonstrates that the generated reviews match or exceed manual quality, offering broad applicability across research fields without requiring users' domain knowledge. Applied to propane dehydrogenation (PDH) catalysts, our method swiftly analyzed 343 articles, averaging seconds per article per LLM account, producing comprehensive reviews spanning 35 topics, with extended analysis of 1041 articles providing insights into catalysts' properties. Through multi-layered quality control, we effectively mitigated LLMs' hallucinations, with expert verification confirming accuracy and citation integrity while demonstrating hallucination risks reduced to below 0.5\% with 95\% confidence. Released Windows application enables one-click review generation, enhancing research productivity and literature recommendation efficiency while setting the stage for broader scientific explorations.
comment: 21 pages, 5 figures, 1 tables Code: https://github.com/TJU-ECAT-AI/AutomaticReviewGeneration Data: https://github.com/TJU-ECAT-AI/AutomaticReviewGenerationData This research has been invited for a Short Oral presentation at the 18th ICC - International Congress on Catalysis, taking place in Lyon, France from July 14-19, 2024
♻ ☆ A Unifying Information-theoretic Perspective on Evaluating Generative Models
Considering the difficulty of interpreting generative model output, there is significant current research focused on determining meaningful evaluation metrics. Several recent approaches utilize "precision" and "recall," borrowed from the classification domain, to individually quantify the output fidelity (realism) and output diversity (representation of the real data variation), respectively. With the increase in metric proposals, there is a need for a unifying perspective, allowing for easier comparison and clearer explanation of their benefits and drawbacks. To this end, we unify a class of kth-nearest-neighbors (kNN)-based metrics under an information-theoretic lens using approaches from kNN density estimation. Additionally, we propose a tri-dimensional metric composed of Precision Cross-Entropy (PCE), Recall Cross-Entropy (RCE), and Recall Entropy (RE), which separately measure fidelity and two distinct aspects of diversity, inter- and intra-class. Our domain-agnostic metric, derived from the information-theoretic concepts of entropy and cross-entropy, can be dissected for both sample- and mode-level analysis. Our detailed experimental results demonstrate the sensitivity of our metric components to their respective qualities and reveal undesirable behaviors of other metrics.
♻ ☆ Enhancing Skin Disease Diagnosis: Interpretable Visual Concept Discovery with SAM WACV 2025
Current AI-assisted skin image diagnosis has achieved dermatologist-level performance in classifying skin cancer, driven by rapid advancements in deep learning architectures. However, unlike traditional vision tasks, skin images in general present unique challenges due to the limited availability of well-annotated datasets, complex variations in conditions, and the necessity for detailed interpretations to ensure patient safety. Previous segmentation methods have sought to reduce image noise and enhance diagnostic performance, but these techniques require fine-grained, pixel-level ground truth masks for training. In contrast, with the rise of foundation models, the Segment Anything Model (SAM) has been introduced to facilitate promptable segmentation, enabling the automation of the segmentation process with simple yet effective prompts. Efforts applying SAM predominantly focus on dermatoscopy images, which present more easily identifiable lesion boundaries than clinical photos taken with smartphones. This limitation constrains the practicality of these approaches to real-world applications. To overcome the challenges posed by noisy clinical photos acquired via non-standardized protocols and to improve diagnostic accessibility, we propose a novel Cross-Attentive Fusion framework for interpretable skin lesion diagnosis. Our method leverages SAM to generate visual concepts for skin diseases using prompts, integrating local visual concepts with global image features to enhance model performance. Extensive evaluation on two skin disease datasets demonstrates our proposed method's effectiveness on lesion diagnosis and interpretability.
comment: This paper is accepted by WACV 2025
♻ ☆ FlowDock: Geometric Flow Matching for Generative Protein-Ligand Docking and Affinity Prediction
Powerful generative AI models of protein-ligand structure have recently been proposed, but few of these methods support both flexible protein-ligand docking and affinity estimation. Of those that do, none can directly model multiple binding ligands concurrently or have been rigorously benchmarked on pharmacologically relevant drug targets, hindering their widespread adoption in drug discovery efforts. In this work, we propose FlowDock, the first deep geometric generative model based on conditional flow matching that learns to directly map unbound (apo) structures to their bound (holo) counterparts for an arbitrary number of binding ligands. Furthermore, FlowDock provides predicted structural confidence scores and binding affinity values with each of its generated protein-ligand complex structures, enabling fast virtual screening of new (multi-ligand) drug targets. For the well-known PoseBusters Benchmark dataset, FlowDock outperforms single-sequence AlphaFold 3 with a 51% blind docking success rate using unbound (apo) protein input structures and without any information derived from multiple sequence alignments, and for the challenging new DockGen-E dataset, FlowDock outperforms single-sequence AlphaFold 3 and matches single-sequence Chai-1 for binding pocket generalization. Additionally, in the ligand category of the 16th community-wide Critical Assessment of Techniques for Structure Prediction (CASP16), FlowDock ranked among the top-5 methods for pharmacological binding affinity estimation across 140 protein-ligand complexes, demonstrating the efficacy of its learned representations in virtual screening. Source code, data, and pre-trained models are available at https://github.com/BioinfoMachineLearning/FlowDock.
comment: 10 pages, 2 tables, 2 algorithms, 7 figures. Code, data, pre-trained models, and baseline method predictions are available at https://github.com/BioinfoMachineLearning/FlowDock
♻ ☆ Multimodal-to-Text Prompt Engineering in Large Language Models Using Feature Embeddings for GNSS Interference Characterization
Large language models (LLMs) are advanced AI systems applied across various domains, including NLP, information retrieval, and recommendation systems. Despite their adaptability and efficiency, LLMs have not been extensively explored for signal processing tasks, particularly in the domain of global navigation satellite system (GNSS) interference monitoring. GNSS interference monitoring is essential to ensure the reliability of vehicle localization on roads, a critical requirement for numerous applications. However, GNSS-based positioning is vulnerable to interference from jamming devices, which can compromise its accuracy. The primary objective is to identify, classify, and mitigate these interferences. Interpreting GNSS snapshots and the associated interferences presents significant challenges due to the inherent complexity, including multipath effects, diverse interference types, varying sensor characteristics, and satellite constellations. In this paper, we extract features from a large GNSS dataset and employ LLaVA to retrieve relevant information from an extensive knowledge base. We employ prompt engineering to interpret the interferences and environmental factors, and utilize t-SNE to analyze the feature embeddings. Our findings demonstrate that the proposed method is capable of visual and logical reasoning within the GNSS context. Furthermore, our pipeline outperforms state-of-the-art machine learning models in interference classification tasks.
♻ ☆ PASS: Presentation Automation for Slide Generation and Speech
In today's fast-paced world, effective presentations have become an essential tool for communication in both online and offline meetings. The crafting of a compelling presentation requires significant time and effort, from gathering key insights to designing slides that convey information clearly and concisely. However, despite the wealth of resources available, people often find themselves manually extracting crucial points, analyzing data, and organizing content in a way that ensures clarity and impact. Furthermore, a successful presentation goes beyond just the slides; it demands rehearsal and the ability to weave a captivating narrative to fully engage the audience. Although there has been some exploration of automating document-to-slide generation, existing research is largely centered on converting research papers. In addition, automation of the delivery of these presentations has yet to be addressed. We introduce PASS, a pipeline used to generate slides from general Word documents, going beyond just research papers, which also automates the oral delivery of the generated slides. PASS analyzes user documents to create a dynamic, engaging presentation with an AI-generated voice. Additionally, we developed an LLM-based evaluation metric to assess our pipeline across three critical dimensions of presentations: relevance, coherence, and redundancy. The data and codes are available at https://github.com/AggarwalTushar/PASS.
♻ ☆ Enhancing Novel Object Detection via Cooperative Foundational Models WACV 2025
In this work, we address the challenging and emergent problem of novel object detection (NOD), focusing on the accurate detection of both known and novel object categories during inference. Traditional object detection algorithms are inherently closed-set, limiting their capability to handle NOD. We present a novel approach to transform existing closed-set detectors into open-set detectors. This transformation is achieved by leveraging the complementary strengths of pre-trained foundational models, specifically CLIP and SAM, through our cooperative mechanism. Furthermore, by integrating this mechanism with state-of-the-art open-set detectors such as GDINO, we establish new benchmarks in object detection performance. Our method achieves 17.42 mAP in novel object detection and 42.08 mAP for known objects on the challenging LVIS dataset. Adapting our approach to the COCO OVD split, we surpass the current state-of-the-art by a margin of 7.2 $ \text{AP}_{50} $ for novel classes. Our code is available at https://rohit901.github.io/coop-foundation-models/ .
comment: Accepted at WACV 2025
♻ ☆ Contextual Evaluation of Large Language Models for Classifying Tropical and Infectious Diseases NeurIPS 2024
While large language models (LLMs) have shown promise for medical question answering, there is limited work focused on tropical and infectious disease-specific exploration. We build on an opensource tropical and infectious diseases (TRINDs) dataset, expanding it to include demographic and semantic clinical and consumer augmentations yielding 11000+ prompts. We evaluate LLM performance on these, comparing generalist and medical LLMs, as well as LLM outcomes to human experts. We demonstrate through systematic experimentation, the benefit of contextual information such as demographics, location, gender, risk factors for optimal LLM response. Finally we develop a prototype of TRINDs-LM, a research tool that provides a playground to navigate how context impacts LLM outputs for health.
comment: Accepted at 2 NeurIPS 2024 workshops: Generative AI for Health Workshop and Workshop on Advancements In Medical Foundation Models: Explainability, Robustness, Security, and Beyond
Robotics 27
☆ Applying General Turn-taking Models to Conversational Human-Robot Interaction
Turn-taking is a fundamental aspect of conversation, but current Human-Robot Interaction (HRI) systems often rely on simplistic, silence-based models, leading to unnatural pauses and interruptions. This paper investigates, for the first time, the application of general turn-taking models, specifically TurnGPT and Voice Activity Projection (VAP), to improve conversational dynamics in HRI. These models are trained on human-human dialogue data using self-supervised learning objectives, without requiring domain-specific fine-tuning. We propose methods for using these models in tandem to predict when a robot should begin preparing responses, take turns, and handle potential interruptions. We evaluated the proposed system in a within-subject study against a traditional baseline system, using the Furhat robot with 39 adults in a conversational setting, in combination with a large language model for autonomous response generation. The results show that participants significantly prefer the proposed system, and it significantly reduces response delays and interruptions.
comment: Accepted at HRI 2025 (the IEEE/ACM International Conference on Human-Robot Interaction)
☆ A Reinforcement Learning Approach to Quiet and Safe UAM Traffic Management
Urban air mobility (UAM) is a transformative system that operates various small aerial vehicles in urban environments to reshape urban transportation. However, integrating UAM into existing urban environments presents a variety of complex challenges. Recent analyses of UAM's operational constraints highlight aircraft noise and system safety as key hurdles to UAM system implementation. Future UAM air traffic management schemes must ensure that the system is both quiet and safe. We propose a multi-agent reinforcement learning approach to manage UAM traffic, aiming at both vertical separation assurance and noise mitigation. Through extensive training, the reinforcement learning agent learns to balance the two primary objectives by employing altitude adjustments in a multi-layer UAM network. The results reveal the tradeoffs among noise impact, traffic congestion, and separation. Overall, our findings demonstrate the potential of reinforcement learning in mitigating UAM's noise impact while maintaining safe separation using altitude adjustments
comment: Paper presented at SciTech 2025
☆ When Uncertainty Leads to Unsafety: Empirical Insights into the Role of Uncertainty in Unmanned Aerial Vehicle Safety
Despite the recent developments in obstacle avoidance and other safety features, autonomous Unmanned Aerial Vehicles (UAVs) continue to face safety challenges. No previous work investigated the relationship between the behavioral uncertainty of a UAV and the unsafety of its flight. By quantifying uncertainty, it is possible to develop a predictor for unsafety, which acts as a flight supervisor. We conducted a large-scale empirical investigation of safety violations using PX4-Autopilot, an open-source UAV software platform. Our dataset of over 5,000 simulated flights, created to challenge obstacle avoidance, allowed us to explore the relation between uncertain UAV decisions and safety violations: up to 89% of unsafe UAV states exhibit significant decision uncertainty, and up to 74% of uncertain decisions lead to unsafe states. Based on these findings, we implemented Superialist (Supervising Autonomous Aerial Vehicles), a runtime uncertainty detector based on autoencoders, the state-of-the-art technology for anomaly detection. Superialist achieved high performance in detecting uncertain behaviors with up to 96% precision and 93% recall. Despite the observed performance degradation when using the same approach for predicting unsafety (up to 74% precision and 87% recall), Superialist enabled early prediction of unsafe states up to 50 seconds in advance.
comment: 36 pages
☆ SLC$^2$-SLAM: Semantic-guided Loop Closure with Shared Latent Code for NeRF SLAM
Targeting the notorious cumulative drift errors in NeRF SLAM, we propose a Semantic-guided Loop Closure with Shared Latent Code, dubbed SLC$^2$-SLAM. Especially, we argue that latent codes stored in many NeRF SLAM systems are not fully exploited, as they are only used for better reconstruction. In this paper, we propose a simple yet effective way to detect potential loops using the same latent codes as local features. To further improve the loop detection performance, we use the semantic information, which are also decoded from the same latent codes to guide the aggregation of local features. Finally, with the potential loops detected, we close them with a graph optimization followed by bundle adjustment to refine both the estimated poses and the reconstructed scene. To evaluate the performance of our SLC$^2$-SLAM, we conduct extensive experiments on Replica and ScanNet datasets. Our proposed semantic-guided loop closure significantly outperforms the pre-trained NetVLAD and ORB combined with Bag-of-Words, which are used in all the other NeRF SLAM with loop closure. As a result, our SLC$^2$-SLAM also demonstrated better tracking and reconstruction performance, especially in larger scenes with more loops, like ScanNet.
comment: 8 pages, 5 figures, 4 tables
☆ Task Allocation in Mobile Robot Fleets: A review
Mobile robot fleets are currently used in different scenarios such as medical environments or logistics. The management of these systems provides different challenges that vary from the control of the movement of each robot to the allocation of tasks to be performed. Task Allocation (TA) problem is a key topic for the proper management of mobile robot fleets to ensure the minimization of energy consumption and quantity of necessary robots. Solutions on this aspect are essential to reach economic and environmental sustainability of robot fleets, mainly in industry applications such as warehouse logistics. The minimization of energy consumption introduces TA problem as an optimization issue which has been treated in recent studies. This work focuses on the analysis of current trends in solving TA of mobile robot fleets. Main TA optimization algorithms are presented, including novel methods based on Artificial Intelligence (AI). Additionally, this work showcases most important results extracted from simulations, including frameworks utilized for the development of the simulations. Finally, some conclusions are obtained from the analysis to target on gaps that must be treated in the future.
☆ GS-LIVO: Real-Time LiDAR, Inertial, and Visual Multi-sensor Fused Odometry with Gaussian Mapping
In recent years, 3D Gaussian splatting (3D-GS) has emerged as a novel scene representation approach. However, existing vision-only 3D-GS methods often rely on hand-crafted heuristics for point-cloud densification and face challenges in handling occlusions and high GPU memory and computation consumption. LiDAR-Inertial-Visual (LIV) sensor configuration has demonstrated superior performance in localization and dense mapping by leveraging complementary sensing characteristics: rich texture information from cameras, precise geometric measurements from LiDAR, and high-frequency motion data from IMU. Inspired by this, we propose a novel real-time Gaussian-based simultaneous localization and mapping (SLAM) system. Our map system comprises a global Gaussian map and a sliding window of Gaussians, along with an IESKF-based odometry. The global Gaussian map consists of hash-indexed voxels organized in a recursive octree, effectively covering sparse spatial volumes while adapting to different levels of detail and scales. The Gaussian map is initialized through multi-sensor fusion and optimized with photometric gradients. Our system incrementally maintains a sliding window of Gaussians, significantly reducing GPU computation and memory consumption by only optimizing the map within the sliding window. Moreover, we implement a tightly coupled multi-sensor fusion odometry with an iterative error state Kalman filter (IESKF), leveraging real-time updating and rendering of the Gaussian map. Our system represents the first real-time Gaussian-based SLAM framework deployable on resource-constrained embedded systems, demonstrated on the NVIDIA Jetson Orin NX platform. The framework achieves real-time performance while maintaining robust multi-sensor fusion capabilities. All implementation algorithms, hardware designs, and CAD models will be publicly available.
☆ Application of Deep Reinforcement Learning to UAV Swarming for Ground Surveillance
This paper summarizes in depth the state of the art of aerial swarms, covering both classical and new reinforcement-learning-based approaches for their management. Then, it proposes a hybrid AI system, integrating deep reinforcement learning in a multi-agent centralized swarm architecture. The proposed system is tailored to perform surveillance of a specific area, searching and tracking ground targets, for security and law enforcement applications. The swarm is governed by a central swarm controller responsible for distributing different search and tracking tasks among the cooperating UAVs. Each UAV agent is then controlled by a collection of cooperative sub-agents, whose behaviors have been trained using different deep reinforcement learning models, tailored for the different task types proposed by the swarm controller. More specifically, proximal policy optimization (PPO) algorithms were used to train the agents' behavior. In addition, several metrics to assess the performance of the swarm in this application were defined. The results obtained through simulation show that our system searches the operation area effectively, acquires the targets in a reasonable time, and is capable of tracking them continuously and consistently.
☆ Self-Organizing Edge Computing Distribution Framework for Visual SLAM
Localization within a known environment is a crucial capability for mobile robots. Simultaneous Localization and Mapping (SLAM) is a prominent solution to this problem. SLAM is a framework that consists of a diverse set of computational tasks ranging from real-time tracking to computation-intensive map optimization. This combination can present a challenge for resource-limited mobile robots. Previously, edge-assisted SLAM methods have demonstrated promising real-time execution capabilities by offloading heavy computations while performing real-time tracking onboard. However, the common approach of utilizing a client-server architecture for offloading is sensitive to server and network failures. In this article, we propose a novel edge-assisted SLAM framework capable of self-organizing fully distributed SLAM execution across a network of devices or functioning on a single device without connectivity. The architecture consists of three layers and is designed to be device-agnostic, resilient to network failures, and minimally invasive to the core SLAM system. We have implemented and demonstrated the framework for monocular ORB SLAM3 and evaluated it in both fully distributed and standalone SLAM configurations against the ORB SLAM3. The experiment results demonstrate that the proposed design matches the accuracy and resource utilization of the monolithic approach while enabling collaborative execution.
comment: 8 pages, 5 figures
☆ Image-to-Force Estimation for Soft Tissue Interaction in Robotic-Assisted Surgery Using Structured Light
For Minimally Invasive Surgical (MIS) robots, accurate haptic interaction force feedback is essential for ensuring the safety of interacting with soft tissue. However, most existing MIS robotic systems cannot facilitate direct measurement of the interaction force with hardware sensors due to space limitations. This letter introduces an effective vision-based scheme that utilizes a One-Shot structured light projection with a designed pattern on soft tissue coupled with haptic information processing through a trained image-to-force neural network. The images captured from the endoscopic stereo camera are analyzed to reconstruct high-resolution 3D point clouds for soft tissue deformation. Based on this, a modified PointNet-based force estimation method is proposed, which excels in representing the complex mechanical properties of soft tissue. Numerical force interaction experiments are conducted on three silicon materials with different stiffness. The results validate the effectiveness of the proposed scheme.
☆ GOTLoc: General Outdoor Text-based Localization Using Scene Graph Retrieval with OpenStreetMap
We propose GOTLoc, a robust localization method capable of operating even in outdoor environments where GPS signals are unavailable. The method achieves this robust localization by leveraging comparisons between scene graphs generated from text descriptions and maps. Existing text-based localization studies typically represent maps as point clouds and identify the most similar scenes by comparing embeddings of text and point cloud data. However, point cloud maps have limited scalability as it is impractical to pre-generate maps for all outdoor spaces. Furthermore, their large data size makes it challenging to store and utilize them directly on actual robots. To address these issues, GOTLoc leverages compact data structures, such as scene graphs, to store spatial information, enabling individual robots to carry and utilize large amounts of map data. Additionally, by utilizing publicly available map data, such as OpenStreetMap, which provides global information on outdoor spaces, we eliminate the need for additional effort to create custom map data. For performance evaluation, we utilized the KITTI360Pose dataset in conjunction with corresponding OpenStreetMap data to compare the proposed method with existing approaches. Our results demonstrate that the proposed method achieves accuracy comparable to algorithms relying on point cloud maps. Moreover, in city-scale tests, GOTLoc required significantly less storage compared to point cloud-based methods and completed overall processing within a few seconds, validating its applicability to real-world robotics. Our code is available at https://github.com/donghwijung/GOTLoc.
☆ LAMS: LLM-Driven Automatic Mode Switching for Assistive Teleoperation
Teleoperating high degrees-of-freedom (DoF) robotic manipulators via low-DoF controllers like joysticks often requires frequent switching between control modes, where each mode maps controller movements to specific robot actions. Manually performing this frequent switching can make teleoperation cumbersome and inefficient. On the other hand, existing automatic mode-switching solutions, such as heuristic-based or learning-based methods, are often task-specific and lack generalizability. In this paper, we introduce LLM-Driven Automatic Mode Switching (LAMS), a novel approach that leverages Large Language Models (LLMs) to automatically switch control modes based on task context. Unlike existing methods, LAMS requires no prior task demonstrations and incrementally improves by integrating user-generated mode-switching examples. We validate LAMS through an ablation study and a user study with 10 participants on complex, long-horizon tasks, demonstrating that LAMS effectively reduces manual mode switches, is preferred over alternative methods, and improves performance over time. The project website with supplementary materials is at https://lams-assistance.github.io/.
☆ Chance-Constrained Sampling-Based MPC for Collision Avoidance in Uncertain Dynamic Environments
Navigating safely in dynamic and uncertain environments is challenging due to uncertainties in perception and motion. This letter presents C2U-MPPI, a robust sampling-based Model Predictive Control (MPC) framework that addresses these challenges by leveraging the Unscented Model Predictive Path Integral (U-MPPI) control strategy with integrated probabilistic chance constraints, ensuring more reliable and efficient navigation under uncertainty. Unlike gradient-based MPC methods, our approach (i) avoids linearization of system dynamics and directly applies non-convex and nonlinear chance constraints, enabling more accurate and flexible optimization, and (ii) enhances computational efficiency by reformulating probabilistic constraints into a deterministic form and employing a layered dynamic obstacle representation, enabling real-time handling of multiple obstacles. Extensive experiments in simulated and real-world human-shared environments validate the effectiveness of our algorithm against baseline methods, showcasing its capability to generate feasible trajectories and control inputs that adhere to system dynamics and constraints in dynamic settings, enabled by unscented-based sampling strategy and risk-sensitive trajectory evaluation. A supplementary video is available at: https://youtu.be/FptAhvJlQm8
comment: This paper has 8 pages, 2 figures, 5 tables
☆ A Framework for Dynamic Situational Awareness in Human Robot Teams: An Interview Study
In human-robot teams, human situational awareness is the operator's conscious knowledge of the team's states, actions, plans and their environment. Appropriate human situational awareness is critical to successful human-robot collaboration. In human-robot teaming, it is often assumed that the best and required level of situational awareness is knowing everything at all times. This view is problematic, because what a human needs to know for optimal team performance varies given the dynamic environmental conditions, task context and roles and capabilities of team members. We explore this topic by interviewing 16 participants with active and repeated experience in diverse human-robot teaming applications. Based on analysis of these interviews, we derive a framework explaining the dynamic nature of required situational awareness in human-robot teaming. In addition, we identify a range of factors affecting the dynamic nature of required and actual levels of situational awareness (i.e., dynamic situational awareness), types of situational awareness inefficiencies resulting from gaps between actual and required situational awareness, and their main consequences. We also reveal various strategies, initiated by humans and robots, that assist in maintaining the required situational awareness. Our findings inform the implementation of accurate estimates of dynamic situational awareness and the design of user-adaptive human-robot interfaces. Therefore, this work contributes to the future design of more collaborative and effective human-robot teams.
☆ Unified Few-shot Crack Segmentation and its Precise 3D Automatic Measurement in Concrete Structures
Visual-Spatial Systems has become increasingly essential in concrete crack inspection. However, existing methods often lacks adaptability to diverse scenarios, exhibits limited robustness in image-based approaches, and struggles with curved or complex geometries. To address these limitations, an innovative framework for two-dimensional (2D) crack detection, three-dimensional (3D) reconstruction, and 3D automatic crack measurement was proposed by integrating computer vision technologies and multi-modal Simultaneous localization and mapping (SLAM) in this study. Firstly, building on a base DeepLabv3+ segmentation model, and incorporating specific refinements utilizing foundation model Segment Anything Model (SAM), we developed a crack segmentation method with strong generalization across unfamiliar scenarios, enabling the generation of precise 2D crack masks. To enhance the accuracy and robustness of 3D reconstruction, Light Detection and Ranging (LiDAR) point clouds were utilized together with image data and segmentation masks. By leveraging both image- and LiDAR-SLAM, we developed a multi-frame and multi-modal fusion framework that produces dense, colorized point clouds, effectively capturing crack semantics at a 3D real-world scale. Furthermore, the crack geometric attributions were measured automatically and directly within 3D dense point cloud space, surpassing the limitations of conventional 2D image-based measurements. This advancement makes the method suitable for structural components with curved and complex 3D geometries. Experimental results across various concrete structures highlight the significant improvements and unique advantages of the proposed method, demonstrating its effectiveness, accuracy, and robustness in real-world applications.
☆ Combining Movement Primitives with Contraction Theory
This paper presents a modular framework for motion planning using movement primitives. Central to the approach is Contraction Theory, a modular stability tool for nonlinear dynamical systems. The approach extends prior methods by achieving parallel and sequential combinations of both discrete and rhythmic movements, while enabling independent modulation of each movement. This modular framework enables a divide-and-conquer strategy to simplify the programming of complex robot motion planning. Simulation examples illustrate the flexibility and versatility of the framework, highlighting its potential to address diverse challenges in robot motion planning.
comment: 8 pages, 4 figures, submitted to Robotics and Automation Letters (RA-L) for review
☆ Estimation-Aware Trajectory Optimization with Set-Valued Measurement Uncertainties
In this paper, we present an optimization-based framework for generating estimation-aware trajectories in scenarios where measurement (output) uncertainties are state-dependent and set-valued. The framework leverages the concept of regularity for set-valued output maps. Specifically, we demonstrate that, for output-regular maps, one can utilize a set-valued observability measure that is concave with respect to finite-horizon state trajectories. By maximizing this measure, optimized estimation-aware trajectories can be designed for a broad class of systems, including those with locally linearized dynamics. To illustrate the effectiveness of the proposed approach, we provide a representative example in the context of trajectory planning for vision-based estimation. We present an estimation-aware trajectory for an uncooperative target-tracking problem that uses a machine learning (ML)-based estimation module on an ego-satellite.
comment: 25 pages, 5 figures
☆ Embodied Scene Understanding for Vision Language Models via MetaVQA
Vision Language Models (VLMs) demonstrate significant potential as embodied AI agents for various mobility applications. However, a standardized, closed-loop benchmark for evaluating their spatial reasoning and sequential decision-making capabilities is lacking. To address this, we present MetaVQA: a comprehensive benchmark designed to assess and enhance VLMs' understanding of spatial relationships and scene dynamics through Visual Question Answering (VQA) and closed-loop simulations. MetaVQA leverages Set-of-Mark prompting and top-down view ground-truth annotations from nuScenes and Waymo datasets to automatically generate extensive question-answer pairs based on diverse real-world traffic scenarios, ensuring object-centric and context-rich instructions. Our experiments show that fine-tuning VLMs with the MetaVQA dataset significantly improves their spatial reasoning and embodied scene comprehension in safety-critical simulations, evident not only in improved VQA accuracies but also in emerging safety-aware driving maneuvers. In addition, the learning demonstrates strong transferability from simulation to real-world observation. Code and data will be publicly available at https://metadriverse.github.io/metavqa .
comment: for the project webpage, see https://metadriverse.github.io/metavqa
☆ AutoLoop: Fast Visual SLAM Fine-tuning through Agentic Curriculum Learning
Current visual SLAM systems face significant challenges in balancing computational efficiency with robust loop closure handling. Traditional approaches require careful manual tuning and incur substantial computational overhead, while learning-based methods either lack explicit loop closure capabilities or implement them through computationally expensive methods. We present AutoLoop, a novel approach that combines automated curriculum learning with efficient fine-tuning for visual SLAM systems. Our method employs a DDPG (Deep Deterministic Policy Gradient) agent to dynamically adjust loop closure weights during training, eliminating the need for manual hyperparameter search while significantly reducing the required training steps. The approach pre-computes potential loop closure pairs offline and leverages them through an agent-guided curriculum, allowing the model to adapt efficiently to new scenarios. Experiments conducted on TartanAir for training and validated across multiple benchmarks including KITTI, EuRoC, ICL-NUIM and TUM RGB-D demonstrate that AutoLoop achieves comparable or superior performance while reducing training time by an order of magnitude compared to traditional approaches. AutoLoop provides a practical solution for rapid adaptation of visual SLAM systems, automating the weight tuning process that traditionally requires multiple manual iterations. Our results show that this automated curriculum strategy not only accelerates training but also maintains or improves the model's performance across diverse environmental conditions.
♻ ☆ Real-World Evaluation of two Cooperative Intersection Management Approaches
Cooperative maneuver planning promises to significantly improve traffic efficiency at unsignalized intersections by leveraging connected automated vehicles. Previous works on this topic have been mostly developed for completely automated traffic in a simple simulated environment. In contrast, our previously introduced planning approaches are specifically designed to handle real-world mixed traffic. The two methods are based on multi-scenario prediction and graph-based reinforcement learning, respectively. This is the first study to perform evaluations in a novel mixed traffic simulation framework as well as real-world drives with prototype connected automated vehicles in public traffic. The simulation features the same connected automated driving software stack as deployed on one of the automated vehicles. Our quantitative evaluations show that cooperative maneuver planning achieves a substantial reduction in crossing times and the number of stops. In a realistic environment with few automated vehicles, there are noticeable efficiency gains with only slightly increasing criticality metrics.
comment: M. Klimke and M. B. Mertens are both first authors with equal contribution. 10 pages, 9 figures, 3 tables, submitted to IEEE Intelligent Transportation Systems Magazine
♻ ☆ Learning Low-Dimensional Strain Models of Soft Robots by Looking at the Evolution of Their Shape with Application to Model-Based Control
Obtaining dynamic models of continuum soft robots is central to the analysis and control of soft robots, and researchers have devoted much attention to the challenge of proposing both data-driven and first-principle solutions. Both avenues have, however, shown their limitations; the former lacks structure and performs poorly outside training data, while the latter requires significant simplifications and extensive expert knowledge to be used in practice. This paper introduces a streamlined method for learning low-dimensional, physics-based models that are both accurate and easy to interpret. We start with an algorithm that uses image data (i.e., shape evolutions) to determine the minimal necessary segments for describing a soft robot's movement. Following this, we apply a dynamic regression and strain sparsification algorithm to identify relevant strains and define the model's dynamics. We validate our approach through simulations with various planar soft manipulators, comparing its performance against other learning strategies, showing that our models are both computationally efficient and 25x more accurate on out-of-training distribution inputs. Finally, we demonstrate that thanks to the capability of the method of generating physically compatible models, the learned models can be straightforwardly combined with model-based control policies.
comment: 8 pages, appearing in Proceedings of the 2025 IEEE 8th International Conference on Soft Robotics (RoboSoft)
♻ ☆ Evaluation of Artificial Intelligence Methods for Lead Time Prediction in Non-Cycled Areas of Automotive Production
The present study examines the effectiveness of applying Artificial Intelligence methods in an automotive production environment to predict unknown lead times in a non-cycle-controlled production area. Data structures are analyzed to identify contextual features and then preprocessed using one-hot encoding. Methods selection focuses on supervised machine learning techniques. In supervised learning methods, regression and classification methods are evaluated. Continuous regression based on target size distribution is not feasible. Classification methods analysis shows that Ensemble Learning and Support Vector Machines are the most suitable. Preliminary study results indicate that gradient boosting algorithms LightGBM, XGBoost, and CatBoost yield the best results. After further testing and extensive hyperparameter optimization, the final method choice is the LightGBM algorithm. Depending on feature availability and prediction interval granularity, relative prediction accuracies of up to 90% can be achieved. Further tests highlight the importance of periodic retraining of AI models to accurately represent complex production processes using the database. The research demonstrates that AI methods can be effectively applied to highly variable production data, adding business value by providing an additional metric for various control tasks while outperforming current non AI-based systems.
♻ ☆ Mind the Error! Detection and Localization of Instruction Errors in Vision-and-Language Navigation IROS'24
Vision-and-Language Navigation in Continuous Environments (VLN-CE) is one of the most intuitive yet challenging embodied AI tasks. Agents are tasked to navigate towards a target goal by executing a set of low-level actions, following a series of natural language instructions. All VLN-CE methods in the literature assume that language instructions are exact. However, in practice, instructions given by humans can contain errors when describing a spatial environment due to inaccurate memory or confusion. Current VLN-CE benchmarks do not address this scenario, making the state-of-the-art methods in VLN-CE fragile in the presence of erroneous instructions from human users. For the first time, we propose a novel benchmark dataset that introduces various types of instruction errors considering potential human causes. This benchmark provides valuable insight into the robustness of VLN systems in continuous environments. We observe a noticeable performance drop (up to -25%) in Success Rate when evaluating the state-of-the-art VLN-CE methods on our benchmark. Moreover, we formally define the task of Instruction Error Detection and Localization, and establish an evaluation protocol on top of our benchmark dataset. We also propose an effective method, based on a cross-modal transformer architecture, that achieves the best performance in error detection and localization, compared to baselines. Surprisingly, our proposed method has revealed errors in the validation set of the two commonly used datasets for VLN-CE, i.e., R2R-CE and RxR-CE, demonstrating the utility of our technique in other tasks. Code and dataset available at https://intelligolabs.github.io/R2RIE-CE
comment: 3 figures, 8 pages. Accepted at IROS'24
♻ ☆ Reward-Driven Automated Curriculum Learning for Interaction-Aware Self-Driving at Unsignalized Intersections
In this work, we present a reward-driven automated curriculum reinforcement learning approach for interaction-aware self-driving at unsignalized intersections, taking into account the uncertainties associated with surrounding vehicles (SVs). These uncertainties encompass the uncertainty of SVs' driving intention and also the quantity of SVs. To deal with this problem, the curriculum set is specifically designed to accommodate a progressively increasing number of SVs. By implementing an automated curriculum selection mechanism, the importance weights are rationally allocated across various curricula, thereby facilitating improved sample efficiency and training outcomes. Furthermore, the reward function is meticulously designed to guide the agent towards effective policy exploration. Thus the proposed framework could proactively address the above uncertainties at unsignalized intersections by employing the automated curriculum learning technique that progressively increases task difficulty, and this ensures safe self-driving through effective interaction with SVs. Comparative experiments are conducted in $Highway\_Env$, and the results indicate that our approach achieves the highest task success rate, attains strong robustness to initialization parameters of the curriculum selection module, and exhibits superior adaptability to diverse situational configurations at unsignalized intersections. Furthermore, the effectiveness of the proposed method is validated using the high-fidelity CARLA simulator.
comment: 8 pages, 6 figures, add grant information, minor textual polishing
♻ ☆ ModCube: Modular, Self-Assembling Cubic Underwater Robot
This paper presents a low-cost, centralized modular underwater robot platform, ModCube, which can be used to study swarm coordination for a wide range of tasks in underwater environments. A ModCube structure consists of multiple ModCube robots. Each robot can move in six DoF with eight thrusters and can be rigidly connected to other ModCube robots with an electromagnet controlled by onboard computer. In this paper, we present a novel method for characterizing and visualizing dynamic behavior, along with four benchmarks to evaluate the morphological performance of the robot. Analysis shows that our ModCube design is desirable for omnidirectional tasks, compared with the configurations widely used by commercial underwater robots. We run real robot experiments in two water tanks to demonstrate the robust control and self-assemble of the proposed system, We also open-source the design and code to facilitate future research.
comment: 8 pages, 8 figures, letter
♻ ☆ RoboHorizon: An LLM-Assisted Multi-View World Model for Long-Horizon Robotic Manipulation
Efficient control in long-horizon robotic manipulation is challenging due to complex representation and policy learning requirements. Model-based visual reinforcement learning (RL) has shown great potential in addressing these challenges but still faces notable limitations, particularly in handling sparse rewards and complex visual features in long-horizon environments. To address these limitations, we propose the Recognize-Sense-Plan-Act (RSPA) pipeline for long-horizon tasks and further introduce RoboHorizon, an LLM-assisted multi-view world model tailored for long-horizon robotic manipulation. In RoboHorizon, pre-trained LLMs generate dense reward structures for multi-stage sub-tasks based on task language instructions, enabling robots to better recognize long-horizon tasks. Keyframe discovery is then integrated into the multi-view masked autoencoder (MAE) architecture to enhance the robot's ability to sense critical task sequences, strengthening its multi-stage perception of long-horizon processes. Leveraging these dense rewards and multi-view representations, a robotic world model is constructed to efficiently plan long-horizon tasks, enabling the robot to reliably act through RL algorithms. Experiments on two representative benchmarks, RLBench and FurnitureBench, show that RoboHorizon outperforms state-of-the-art visual model-based RL methods, achieving a 23.35% improvement in task success rates on RLBench's 4 short-horizon tasks and a 29.23% improvement on 6 long-horizon tasks from RLBench and 3 furniture assembly tasks from FurnitureBench.
comment: Under review
♻ ☆ On the Surprising Effectiveness of Spectrum Clipping in Learning Stable Linear Dynamics
When learning stable linear dynamical systems from data, three important properties are desirable: i) predictive accuracy, ii) provable stability, and iii) computational efficiency. Unconstrained minimization of reconstruction errors leads to high accuracy and efficiency but cannot guarantee stability. Existing methods to remedy this focus on enforcing stability while also ensuring accuracy, but do so only at the cost of increased computation. In this work, we investigate if a straightforward approach can simultaneously offer all three desiderata of learning stable linear systems. Specifically, we consider a post-hoc approach that manipulates the spectrum of the learned system matrix after it is learned in an unconstrained fashion. We call this approach spectrum clipping (SC) as it involves eigen decomposition and subsequent reconstruction of the system matrix after clipping all of its eigenvalues that are larger than one to one (without altering the eigenvectors). Through detailed experiments involving two different applications and publicly available benchmark datasets, we demonstrate that this simple technique can simultaneously learn highly accurate linear systems that are provably stable. Notably, we demonstrate that SC can achieve similar or better performance than strong baselines while being orders-of-magnitude faster. We also show that SC can be readily combined with Koopman operators to learn stable nonlinear dynamics, such as those underlying complex dexterous manipulation skills involving multi-fingered robotic hands. Further, we find that SC can learn stable robot policies even when the training data includes unsuccessful or truncated demonstrations. Our codes and dataset can be found at https://github.com/GT-STAR-Lab/spec_clip.
comment: Under review by L4DC 2025
♻ ☆ Experimental Study on The Effect of Multi-step Deep Reinforcement Learning in POMDPs
Deep Reinforcement Learning (DRL) has made tremendous advances in both simulated and real-world robot control tasks in recent years. This is particularly the case for tasks that can be carefully engineered with a full state representation, and which can then be formulated as a Markov Decision Process (MDP). However, applying DRL strategies designed for MDPs to novel robot control tasks can be challenging, because the available observations may be a partial representation of the state, resulting in a Partially Observable Markov Decision Process (POMDP). This paper considers three popular DRL algorithms, namely Proximal Policy Optimization (PPO), Twin Delayed Deep Deterministic Policy Gradient (TD3), and Soft Actor-Critic (SAC), invented for MDPs, and studies their performance in POMDP scenarios. While prior work has found that SAC and TD3 typically outperform PPO across a broad range of tasks that can be represented as MDPs, we show that this is not always the case, using three representative POMDP environments. Empirical studies show that this is related to multi-step bootstrapping, where multi-step immediate rewards, instead of one-step immediate reward, are used to calculate the target value estimation of an observation and action pair. We identify this by observing that the inclusion of multi-step bootstrapping in TD3 (MTD3) and SAC (MSAC) results in improved robustness in POMDP settings.
Multimedia 3
☆ Multimodal LLMs Can Reason about Aesthetics in Zero-Shot
We present the first study on how Multimodal LLMs' (MLLMs) reasoning ability shall be elicited to evaluate the aesthetics of artworks. To facilitate this investigation, we construct MM-StyleBench, a novel high-quality dataset for benchmarking artistic stylization. We then develop a principled method for human preference modeling and perform a systematic correlation analysis between MLLMs' responses and human preference. Our experiments reveal an inherent hallucination issue of MLLMs in art evaluation, associated with response subjectivity. ArtCoT is proposed, demonstrating that art-specific task decomposition and the use of concrete language boost MLLMs' reasoning ability for aesthetics. Our findings offer valuable insights into MLLMs for art and can benefit a wide range of downstream applications, such as style transfer and artistic image generation. Code available at https://github.com/songrise/MLLM4Art.
comment: WIP, Homepage https://github.com/songrise/MLLM4Art
☆ Multimodal Fake News Video Explanation Generation
Multi-modal explanation involves the assessment of the veracity of a variety of different content, and relies on multiple information modalities to comprehensively consider the relevance and consistency between modalities. Most existing fake news video detection methods focus on improving accuracy while ignoring the importance of providing explanations. In this paper, we propose a novel problem - Fake News Video Explanation (FNVE) - Given a multimodal news containing both video and caption text, we aim to generate natural language explanations to reveal the truth of predictions. To this end, we develop FakeNVE, a new dataset of explanations for truthfully multimodal posts, where each explanation is a natural language (English) sentence describing the attribution of a news thread. We benchmark FakeNVE by using a multimodal transformer-based architecture. Subsequently, a BART-based autoregressive decoder is used as the generator. Empirical results show compelling results for various baselines (applicable to FNVE) across multiple evaluation metrics. We also perform human evaluation on explanation generation, achieving high scores for both adequacy and fluency.
♻ ☆ Multi-modal and Multi-scale Spatial Environment Understanding for Immersive Visual Text-to-Speech AAAI'2025
Visual Text-to-Speech (VTTS) aims to take the environmental image as the prompt to synthesize the reverberant speech for the spoken content. The challenge of this task lies in understanding the spatial environment from the image. Many attempts have been made to extract global spatial visual information from the RGB space of an spatial image. However, local and depth image information are crucial for understanding the spatial environment, which previous works have ignored. To address the issues, we propose a novel multi-modal and multi-scale spatial environment understanding scheme to achieve immersive VTTS, termed M2SE-VTTS. The multi-modal aims to take both the RGB and Depth spaces of the spatial image to learn more comprehensive spatial information, and the multi-scale seeks to model the local and global spatial knowledge simultaneously. Specifically, we first split the RGB and Depth images into patches and adopt the Gemini-generated environment captions to guide the local spatial understanding. After that, the multi-modal and multi-scale features are integrated by the local-aware global spatial understanding. In this way, M2SE-VTTS effectively models the interactions between local and global spatial contexts in the multi-modal spatial environment. Objective and subjective evaluations suggest that our model outperforms the advanced baselines in environmental speech generation. The code and audio samples are available at: https://github.com/AI-S2-Lab/M2SE-VTTS.
comment: 9 pages,2 figures, Accepted by AAAI'2025
Computer Vision and Pattern Recognition 151
☆ DAViD: Modeling Dynamic Affordance of 3D Objects using Pre-trained Video Diffusion Models
Understanding the ability of humans to use objects is crucial for AI to improve daily life. Existing studies for learning such ability focus on human-object patterns (e.g., contact, spatial relation, orientation) in static situations, and learning Human-Object Interaction (HOI) patterns over time (i.e., movement of human and object) is relatively less explored. In this paper, we introduce a novel type of affordance named Dynamic Affordance. For a given input 3D object mesh, we learn dynamic affordance which models the distribution of both (1) human motion and (2) human-guided object pose during interactions. As a core idea, we present a method to learn the 3D dynamic affordance from synthetically generated 2D videos, leveraging a pre-trained video diffusion model. Specifically, we propose a pipeline that first generates 2D HOI videos from the 3D object and then lifts them into 3D to generate 4D HOI samples. Once we generate diverse 4D HOI samples on various target objects, we train our DAViD, where we present a method based on the Low-Rank Adaptation (LoRA) module for pre-trained human motion diffusion model (MDM) and an object pose diffusion model with human pose guidance. Our motion diffusion model is extended for multi-object interactions, demonstrating the advantage of our pipeline with LoRA for combining the concepts of object usage. Through extensive experiments, we demonstrate our DAViD outperforms the baselines in generating human motion with HOIs.
comment: Project Page: https://snuvclab.github.io/david/
☆ MangaNinja: Line Art Colorization with Precise Reference Following
Derived from diffusion models, MangaNinjia specializes in the task of reference-guided line art colorization. We incorporate two thoughtful designs to ensure precise character detail transcription, including a patch shuffling module to facilitate correspondence learning between the reference color image and the target line art, and a point-driven control scheme to enable fine-grained color matching. Experiments on a self-collected benchmark demonstrate the superiority of our model over current solutions in terms of precise colorization. We further showcase the potential of the proposed interactive point control in handling challenging cases, cross-character colorization, multi-reference harmonization, beyond the reach of existing algorithms.
comment: Project page and code: https://johanan528.github.io/MangaNinjia/
☆ Go-with-the-Flow: Motion-Controllable Video Diffusion Models Using Real-Time Warped Noise
Generative modeling aims to transform random noise into structured outputs. In this work, we enhance video diffusion models by allowing motion control via structured latent noise sampling. This is achieved by just a change in data: we pre-process training videos to yield structured noise. Consequently, our method is agnostic to diffusion model design, requiring no changes to model architectures or training pipelines. Specifically, we propose a novel noise warping algorithm, fast enough to run in real time, that replaces random temporal Gaussianity with correlated warped noise derived from optical flow fields, while preserving the spatial Gaussianity. The efficiency of our algorithm enables us to fine-tune modern video diffusion base models using warped noise with minimal overhead, and provide a one-stop solution for a wide range of user-friendly motion control: local object motion control, global camera movement control, and motion transfer. The harmonization between temporal coherence and spatial Gaussianity in our warped noise leads to effective motion control while maintaining per-frame pixel quality. Extensive experiments and user studies demonstrate the advantages of our method, making it a robust and scalable approach for controlling motion in video diffusion models. Video results are available on our webpage: https://vgenai-netflix-eyeline-research.github.io/Go-with-the-Flow/; source code and model checkpoints are available on GitHub: https://github.com/VGenAI-Netflix-Eyeline-Research/Go-with-the-Flow.
☆ Predicting 4D Hand Trajectory from Monocular Videos
We present HaPTIC, an approach that infers coherent 4D hand trajectories from monocular videos. Current video-based hand pose reconstruction methods primarily focus on improving frame-wise 3D pose using adjacent frames rather than studying consistent 4D hand trajectories in space. Despite the additional temporal cues, they generally underperform compared to image-based methods due to the scarcity of annotated video data. To address these issues, we repurpose a state-of-the-art image-based transformer to take in multiple frames and directly predict a coherent trajectory. We introduce two types of lightweight attention layers: cross-view self-attention to fuse temporal information, and global cross-attention to bring in larger spatial context. Our method infers 4D hand trajectories similar to the ground truth while maintaining strong 2D reprojection alignment. We apply the method to both egocentric and allocentric videos. It significantly outperforms existing methods in global trajectory accuracy while being comparable to the state-of-the-art in single-image pose estimation. Project website: https://judyye.github.io/haptic-www
☆ Omni-RGPT: Unifying Image and Video Region-level Understanding via Token Marks
We present Omni-RGPT, a multimodal large language model designed to facilitate region-level comprehension for both images and videos. To achieve consistent region representation across spatio-temporal dimensions, we introduce Token Mark, a set of tokens highlighting the target regions within the visual feature space. These tokens are directly embedded into spatial regions using region prompts (e.g., boxes or masks) and simultaneously incorporated into the text prompt to specify the target, establishing a direct connection between visual and text tokens. To further support robust video understanding without requiring tracklets, we introduce an auxiliary task that guides Token Mark by leveraging the consistency of the tokens, enabling stable region interpretation across the video. Additionally, we introduce a large-scale region-level video instruction dataset (RegVID-300k). Omni-RGPT achieves state-of-the-art results on image and video-based commonsense reasoning benchmarks while showing strong performance in captioning and referring expression comprehension tasks.
comment: Project page: https://miranheo.github.io/omni-rgpt/
☆ GameFactory: Creating New Games with Generative Interactive Videos
Generative game engines have the potential to revolutionize game development by autonomously creating new content and reducing manual workload. However, existing video-based game generation methods fail to address the critical challenge of scene generalization, limiting their applicability to existing games with fixed styles and scenes. In this paper, we present GameFactory, a framework focused on exploring scene generalization in game video generation. To enable the creation of entirely new and diverse games, we leverage pre-trained video diffusion models trained on open-domain video data. To bridge the domain gap between open-domain priors and small-scale game dataset, we propose a multi-phase training strategy that decouples game style learning from action control, preserving open-domain generalization while achieving action controllability. Using Minecraft as our data source, we release GF-Minecraft, a high-quality and diversity action-annotated video dataset for research. Furthermore, we extend our framework to enable autoregressive action-controllable game video generation, allowing the production of unlimited-length interactive game videos. Experimental results demonstrate that GameFactory effectively generates open-domain, diverse, and action-controllable game videos, representing a significant step forward in AI-driven game generation. Our dataset and project page are publicly available at \url{https://vvictoryuki.github.io/gamefactory/}.
☆ Diffusion Adversarial Post-Training for One-Step Video Generation
The diffusion models are widely used for image and video generation, but their iterative generation process is slow and expansive. While existing distillation approaches have demonstrated the potential for one-step generation in the image domain, they still suffer from significant quality degradation. In this work, we propose Adversarial Post-Training (APT) against real data following diffusion pre-training for one-step video generation. To improve the training stability and quality, we introduce several improvements to the model architecture and training procedures, along with an approximated R1 regularization objective. Empirically, our experiments show that our adversarial post-trained model, Seaweed-APT, can generate 2-second, 1280x720, 24fps videos in real time using a single forward evaluation step. Additionally, our model is capable of generating 1024px images in a single step, achieving quality comparable to state-of-the-art methods.
☆ MiniMax-01: Scaling Foundation Models with Lightning Attention
We introduce MiniMax-01 series, including MiniMax-Text-01 and MiniMax-VL-01, which are comparable to top-tier models while offering superior capabilities in processing longer contexts. The core lies in lightning attention and its efficient scaling. To maximize computational capacity, we integrate it with Mixture of Experts (MoE), creating a model with 32 experts and 456 billion total parameters, of which 45.9 billion are activated for each token. We develop an optimized parallel strategy and highly efficient computation-communication overlap techniques for MoE and lightning attention. This approach enables us to conduct efficient training and inference on models with hundreds of billions of parameters across contexts spanning millions of tokens. The context window of MiniMax-Text-01 can reach up to 1 million tokens during training and extrapolate to 4 million tokens during inference at an affordable cost. Our vision-language model, MiniMax-VL-01 is built through continued training with 512 billion vision-language tokens. Experiments on both standard and in-house benchmarks show that our models match the performance of state-of-the-art models like GPT-4o and Claude-3.5-Sonnet while offering 20-32 times longer context window. We publicly release MiniMax-01 at https://github.com/MiniMax-AI.
comment: A technical report from MiniMax. The authors are listed in alphabetical order. We open-sourced our MiniMax-01 at https://github.com/MiniMax-AI
☆ Advancing Semantic Future Prediction through Multimodal Visual Sequence Transformers
Semantic future prediction is important for autonomous systems navigating dynamic environments. This paper introduces FUTURIST, a method for multimodal future semantic prediction that uses a unified and efficient visual sequence transformer architecture. Our approach incorporates a multimodal masked visual modeling objective and a novel masking mechanism designed for multimodal training. This allows the model to effectively integrate visible information from various modalities, improving prediction accuracy. Additionally, we propose a VAE-free hierarchical tokenization process, which reduces computational complexity, streamlines the training pipeline, and enables end-to-end training with high-resolution, multimodal inputs. We validate FUTURIST on the Cityscapes dataset, demonstrating state-of-the-art performance in future semantic segmentation for both short- and mid-term forecasting. We provide the implementation code at https://github.com/Sta8is/FUTURIST .
☆ LayerAnimate: Layer-specific Control for Animation
Animated video separates foreground and background elements into layers, with distinct processes for sketching, refining, coloring, and in-betweening. Existing video generation methods typically treat animation as a monolithic data domain, lacking fine-grained control over individual layers. In this paper, we introduce LayerAnimate, a novel architectural approach that enhances fine-grained control over individual animation layers within a video diffusion model, allowing users to independently manipulate foreground and background elements in distinct layers. To address the challenge of limited layer-specific data, we propose a data curation pipeline that features automated element segmentation, motion-state hierarchical merging, and motion coherence refinement. Through quantitative and qualitative comparisons, and user study, we demonstrate that LayerAnimate outperforms current methods in terms of animation quality, control precision, and usability, making it an ideal tool for both professional animators and amateur enthusiasts. This framework opens up new possibilities for layer-specific animation applications and creative flexibility. Our code is available at https://layeranimate.github.io.
comment: Project page: https://layeranimate.github.io
☆ VINGS-Mono: Visual-Inertial Gaussian Splatting Monocular SLAM in Large Scenes
VINGS-Mono is a monocular (inertial) Gaussian Splatting (GS) SLAM framework designed for large scenes. The framework comprises four main components: VIO Front End, 2D Gaussian Map, NVS Loop Closure, and Dynamic Eraser. In the VIO Front End, RGB frames are processed through dense bundle adjustment and uncertainty estimation to extract scene geometry and poses. Based on this output, the mapping module incrementally constructs and maintains a 2D Gaussian map. Key components of the 2D Gaussian Map include a Sample-based Rasterizer, Score Manager, and Pose Refinement, which collectively improve mapping speed and localization accuracy. This enables the SLAM system to handle large-scale urban environments with up to 50 million Gaussian ellipsoids. To ensure global consistency in large-scale scenes, we design a Loop Closure module, which innovatively leverages the Novel View Synthesis (NVS) capabilities of Gaussian Splatting for loop closure detection and correction of the Gaussian map. Additionally, we propose a Dynamic Eraser to address the inevitable presence of dynamic objects in real-world outdoor scenes. Extensive evaluations in indoor and outdoor environments demonstrate that our approach achieves localization performance on par with Visual-Inertial Odometry while surpassing recent GS/NeRF SLAM methods. It also significantly outperforms all existing methods in terms of mapping and rendering quality. Furthermore, we developed a mobile app and verified that our framework can generate high-quality Gaussian maps in real time using only a smartphone camera and a low-frequency IMU sensor. To the best of our knowledge, VINGS-Mono is the first monocular Gaussian SLAM method capable of operating in outdoor environments and supporting kilometer-scale large scenes.
☆ Can Bayesian Neural Networks Explicitly Model Input Uncertainty?
Inputs to machine learning models can have associated noise or uncertainties, but they are often ignored and not modelled. It is unknown if Bayesian Neural Networks and their approximations are able to consider uncertainty in their inputs. In this paper we build a two input Bayesian Neural Network (mean and standard deviation) and evaluate its capabilities for input uncertainty estimation across different methods like Ensembles, MC-Dropout, and Flipout. Our results indicate that only some uncertainty estimation methods for approximate Bayesian NNs can model input uncertainty, in particular Ensembles and Flipout.
comment: 12 pages, 11 figures, VISAPP 2025 camera ready
☆ LLaVA-ST: A Multimodal Large Language Model for Fine-Grained Spatial-Temporal Understanding
Recent advancements in multimodal large language models (MLLMs) have shown promising results, yet existing approaches struggle to effectively handle both temporal and spatial localization simultaneously. This challenge stems from two key issues: first, incorporating spatial-temporal localization introduces a vast number of coordinate combinations, complicating the alignment of linguistic and visual coordinate representations; second, encoding fine-grained temporal and spatial information during video feature compression is inherently difficult. To address these issues, we propose LLaVA-ST, a MLLM for fine-grained spatial-temporal multimodal understanding. In LLaVA-ST, we propose Language-Aligned Positional Embedding, which embeds the textual coordinate special token into the visual space, simplifying the alignment of fine-grained spatial-temporal correspondences. Additionally, we design the Spatial-Temporal Packer, which decouples the feature compression of temporal and spatial resolutions into two distinct point-to-region attention processing streams. Furthermore, we propose ST-Align dataset with 4.3M training samples for fine-grained spatial-temporal multimodal understanding. With ST-align, we present a progressive training pipeline that aligns the visual and textual feature through sequential coarse-to-fine stages.Additionally, we introduce an ST-Align benchmark to evaluate spatial-temporal interleaved fine-grained understanding tasks, which include Spatial-Temporal Video Grounding (STVG) , Event Localization and Captioning (ELC) and Spatial Video Grounding (SVG). LLaVA-ST achieves outstanding performance on 11 benchmarks requiring fine-grained temporal, spatial, or spatial-temporal interleaving multimodal understanding. Our code, data and benchmark will be released at Our code, data and benchmark will be released at https://github.com/appletea233/LLaVA-ST .
☆ SmartEraser: Remove Anything from Images using Masked-Region Guidance
Object removal has so far been dominated by the mask-and-inpaint paradigm, where the masked region is excluded from the input, leaving models relying on unmasked areas to inpaint the missing region. However, this approach lacks contextual information for the masked area, often resulting in unstable performance. In this work, we introduce SmartEraser, built with a new removing paradigm called Masked-Region Guidance. This paradigm retains the masked region in the input, using it as guidance for the removal process. It offers several distinct advantages: (a) it guides the model to accurately identify the object to be removed, preventing its regeneration in the output; (b) since the user mask often extends beyond the object itself, it aids in preserving the surrounding context in the final result. Leveraging this new paradigm, we present Syn4Removal, a large-scale object removal dataset, where instance segmentation data is used to copy and paste objects onto images as removal targets, with the original images serving as ground truths. Experimental results demonstrate that SmartEraser significantly outperforms existing methods, achieving superior performance in object removal, especially in complex scenes with intricate compositions.
comment: Project at: https://longtaojiang.github.io/smarteraser.github.io/
☆ AI Driven Water Segmentation with deep learning models for Enhanced Flood Monitoring
Flooding is a major natural hazard causing significant fatalities and economic losses annually, with increasing frequency due to climate change. Rapid and accurate flood detection and monitoring are crucial for mitigating these impacts. This study compares the performance of three deep learning models UNet, ResNet, and DeepLabv3 for pixelwise water segmentation to aid in flood detection, utilizing images from drones, in field observations, and social media. This study involves creating a new dataset that augments wellknown benchmark datasets with flood-specific images, enhancing the robustness of the models. The UNet, ResNet, and DeepLab v3 architectures are tested to determine their effectiveness in various environmental conditions and geographical locations, and the strengths and limitations of each model are also discussed here, providing insights into their applicability in different scenarios by predicting image segmentation masks. This fully automated approach allows these models to isolate flooded areas in images, significantly reducing processing time compared to traditional semi-automated methods. The outcome of this study is to predict segmented masks for each image effected by a flood disaster and the validation accuracy of these models. This methodology facilitates timely and continuous flood monitoring, providing vital data for emergency response teams to reduce loss of life and economic damages. It offers a significant reduction in the time required to generate flood maps, cutting down the manual processing time. Additionally, we present avenues for future research, including the integration of multimodal data sources and the development of robust deep learning architectures tailored specifically for flood detection tasks. Overall, our work contributes to the advancement of flood management strategies through innovative use of deep learning technologies.
comment: 8 pages, 6 figures
☆ Towards an End-to-End (E2E) Adversarial Learning and Application in the Physical World
The traditional learning process of patch-based adversarial attacks, conducted in the digital domain and then applied in the physical domain (e.g., via printed stickers), may suffer from reduced performance due to adversarial patches' limited transferability from the digital domain to the physical domain. Given that previous studies have considered using projectors to apply adversarial attacks, we raise the following question: can adversarial learning (i.e., patch generation) be performed entirely in the physical domain with a projector? In this work, we propose the Physical-domain Adversarial Patch Learning Augmentation (PAPLA) framework, a novel end-to-end (E2E) framework that converts adversarial learning from the digital domain to the physical domain using a projector. We evaluate PAPLA across multiple scenarios, including controlled laboratory settings and realistic outdoor environments, demonstrating its ability to ensure attack success compared to conventional digital learning-physical application (DL-PA) methods. We also analyze the impact of environmental factors, such as projection surface color, projector strength, ambient light, distance, and angle of the target object relative to the camera, on the effectiveness of projected patches. Finally, we demonstrate the feasibility of the attack against a parked car and a stop sign in a real-world outdoor environment. Our results show that under specific conditions, E2E adversarial learning in the physical domain eliminates the transferability issue and ensures evasion by object detectors. Finally, we provide insights into the challenges and opportunities of applying adversarial learning in the physical domain and explain where such an approach is more effective than using a sticker.
☆ Continual Deep Active Learning for Medical Imaging: Replay-Base Architecture for Context Adaptation
Deep Learning for medical imaging faces challenges in adapting and generalizing to new contexts. Additionally, it often lacks sufficient labeled data for specific tasks requiring significant annotation effort. Continual Learning (CL) tackles adaptability and generalizability by enabling lifelong learning from a data stream while mitigating forgetting of previously learned knowledge. Active Learning (AL) reduces the number of required annotations for effective training. This work explores both approaches (CAL) to develop a novel framework for robust medical image analysis. Based on the automatic recognition of shifts in image characteristics, Replay-Base Architecture for Context Adaptation (RBACA) employs a CL rehearsal method to continually learn from diverse contexts, and an AL component to select the most informative instances for annotation. A novel approach to evaluate CAL methods is established using a defined metric denominated IL-Score, which allows for the simultaneous assessment of transfer learning, forgetting, and final model performance. We show that RBACA works in domain and class-incremental learning scenarios, by assessing its IL-Score on the segmentation and diagnosis of cardiac images. The results show that RBACA outperforms a baseline framework without CAL, and a state-of-the-art CAL method across various memory sizes and annotation budgets. Our code is available in https://github.com/RuiDaniel/RBACA .
☆ A Feature-Level Ensemble Model for COVID-19 Identification in CXR Images using Choquet Integral and Differential Evolution Optimization
The COVID-19 pandemic has profoundly impacted billions globally. It challenges public health and healthcare systems due to its rapid spread and severe respiratory effects. An effective strategy to mitigate the COVID-19 pandemic involves integrating testing to identify infected individuals. While RT-PCR is considered the gold standard for diagnosing COVID-19, it has some limitations such as the risk of false negatives. To address this problem, this paper introduces a novel Deep Learning Diagnosis System that integrates pre-trained Deep Convolutional Neural Networks (DCNNs) within an ensemble learning framework to achieve precise identification of COVID-19 cases from Chest X-ray (CXR) images. We combine feature vectors from the final hidden layers of pre-trained DCNNs using the Choquet integral to capture interactions between different DCNNs that a linear approach cannot. We employed Sugeno-$\lambda$ measure theory to derive fuzzy measures for subsets of networks to enable aggregation. We utilized Differential Evolution to estimate fuzzy densities. We developed a TensorFlow-based layer for Choquet operation to facilitate efficient aggregation, due to the intricacies involved in aggregating feature vectors. Experimental results on the COVIDx dataset show that our ensemble model achieved 98\% accuracy in three-class classification and 99.50\% in binary classification, outperforming its components-DenseNet-201 (97\% for three-class, 98.75\% for binary), Inception-v3 (96.25\% for three-class, 98.50\% for binary), and Xception (94.50\% for three-class, 98\% for binary)-and surpassing many previous methods.
☆ Efficient Deep Learning-based Forward Solvers for Brain Tumor Growth Models
Glioblastoma, a highly aggressive brain tumor, poses major challenges due to its poor prognosis and high morbidity rates. Partial differential equation-based models offer promising potential to enhance therapeutic outcomes by simulating patient-specific tumor behavior for improved radiotherapy planning. However, model calibration remains a bottleneck due to the high computational demands of optimization methods like Monte Carlo sampling and evolutionary algorithms. To address this, we recently introduced an approach leveraging a neural forward solver with gradient-based optimization to significantly reduce calibration time. This approach requires a highly accurate and fully differentiable forward model. We investigate multiple architectures, including (i) an enhanced TumorSurrogate, (ii) a modified nnU-Net, and (iii) a 3D Vision Transformer (ViT). The optimized TumorSurrogate achieved the best overall results, excelling in both tumor outline matching and voxel-level prediction of tumor cell concentration. It halved the MSE relative to the baseline model and achieved the highest Dice score across all tumor cell concentration thresholds. Our study demonstrates significant enhancement in forward solver performance and outlines important future research directions.
☆ FramePainter: Endowing Interactive Image Editing with Video Diffusion Priors
Interactive image editing allows users to modify images through visual interaction operations such as drawing, clicking, and dragging. Existing methods construct such supervision signals from videos, as they capture how objects change with various physical interactions. However, these models are usually built upon text-to-image diffusion models, so necessitate (i) massive training samples and (ii) an additional reference encoder to learn real-world dynamics and visual consistency. In this paper, we reformulate this task as an image-to-video generation problem, so that inherit powerful video diffusion priors to reduce training costs and ensure temporal consistency. Specifically, we introduce FramePainter as an efficient instantiation of this formulation. Initialized with Stable Video Diffusion, it only uses a lightweight sparse control encoder to inject editing signals. Considering the limitations of temporal attention in handling large motion between two frames, we further propose matching attention to enlarge the receptive field while encouraging dense correspondence between edited and source image tokens. We highlight the effectiveness and efficiency of FramePainter across various of editing signals: it domainantly outperforms previous state-of-the-art methods with far less training data, achieving highly seamless and coherent editing of images, \eg, automatically adjust the reflection of the cup. Moreover, FramePainter also exhibits exceptional generalization in scenarios not present in real-world videos, \eg, transform the clownfish into shark-like shape. Our code will be available at https://github.com/YBYBZhang/FramePainter.
comment: Code: https://github.com/YBYBZhang/FramePainter
☆ EmoNeXt: an Adapted ConvNeXt for Facial Emotion Recognition SP
Facial expressions play a crucial role in human communication serving as a powerful and impactful means to express a wide range of emotions. With advancements in artificial intelligence and computer vision, deep neural networks have emerged as effective tools for facial emotion recognition. In this paper, we propose EmoNeXt, a novel deep learning framework for facial expression recognition based on an adapted ConvNeXt architecture network. We integrate a Spatial Transformer Network (STN) to focus on feature-rich regions of the face and Squeeze-and-Excitation blocks to capture channel-wise dependencies. Moreover, we introduce a self-attention regularization term, encouraging the model to generate compact feature vectors. We demonstrate the superiority of our model over existing state-of-the-art deep learning models on the FER2013 dataset regarding emotion classification accuracy.
comment: 6 pages, 5 figures and 2 tables. 2023 IEEE 25th International Workshop on Multimedia Signal Processing (MMSP), Poitiers, France
Self-supervised Deep Hyperspectral Inpainting with the Plug and Play and Deep Image Prior Models
Hyperspectral images are typically composed of hundreds of narrow and contiguous spectral bands, each containing information regarding the material composition of the imaged scene. However, these images can be affected by various sources of noise, distortions, or data loss, which can significantly degrade their quality and usefulness. This paper introduces a convergent guaranteed algorithm, LRS-PnP-DIP(1-Lip), which successfully addresses the instability issue of DHP that has been reported before. The proposed algorithm extends the successful joint low-rank and sparse model to further exploit the underlying data structures beyond the conventional and sometimes restrictive unions of subspace models. A stability analysis guarantees the convergence of the proposed algorithm under mild assumptions , which is crucial for its application in real-world scenarios. Extensive experiments demonstrate that the proposed solution consistently delivers visually and quantitatively superior inpainting results, establishing state-of-the-art performance.
comment: 31 pages, 9 Figures, 7 Tables. arXiv admin note: text overlap with arXiv:2306.08128
☆ A Critical Synthesis of Uncertainty Quantification and Foundation Models in Monocular Depth Estimation
While recent foundation models have enabled significant breakthroughs in monocular depth estimation, a clear path towards safe and reliable deployment in the real-world remains elusive. Metric depth estimation, which involves predicting absolute distances, poses particular challenges, as even the most advanced foundation models remain prone to critical errors. Since quantifying the uncertainty has emerged as a promising endeavor to address these limitations and enable trustworthy deployment, we fuse five different uncertainty quantification methods with the current state-of-the-art DepthAnythingV2 foundation model. To cover a wide range of metric depth domains, we evaluate their performance on four diverse datasets. Our findings identify fine-tuning with the Gaussian Negative Log-Likelihood Loss (GNLL) as a particularly promising approach, offering reliable uncertainty estimates while maintaining predictive performance and computational efficiency on par with the baseline, encompassing both training and inference time. By fusing uncertainty quantification and foundation models within the context of monocular depth estimation, this paper lays a critical foundation for future research aimed at improving not only model performance but also its explainability. Extending this critical synthesis of uncertainty quantification and foundation models into other crucial tasks, such as semantic segmentation and pose estimation, presents exciting opportunities for safer and more reliable machine vision systems.
☆ CG-MER: A Card Game-based Multimodal dataset for Emotion Recognition
The field of affective computing has seen significant advancements in exploring the relationship between emotions and emerging technologies. This paper presents a novel and valuable contribution to this field with the introduction of a comprehensive French multimodal dataset designed specifically for emotion recognition. The dataset encompasses three primary modalities: facial expressions, speech, and gestures, providing a holistic perspective on emotions. Moreover, the dataset has the potential to incorporate additional modalities, such as Natural Language Processing (NLP) to expand the scope of emotion recognition research. The dataset was curated through engaging participants in card game sessions, where they were prompted to express a range of emotions while responding to diverse questions. The study included 10 sessions with 20 participants (9 females and 11 males). The dataset serves as a valuable resource for furthering research in emotion recognition and provides an avenue for exploring the intricate connections between human emotions and digital technologies.
comment: 8 pages, 2 figures and 4 tables. Sixteenth International Conference on Machine Vision (ICMV 2023), Yerevan, Armenia
☆ D$^2$-DPM: Dual Denoising for Quantized Diffusion Probabilistic Models AAAI2025
Diffusion models have achieved cutting-edge performance in image generation. However, their lengthy denoising process and computationally intensive score estimation network impede their scalability in low-latency and resource-constrained scenarios. Post-training quantization (PTQ) compresses and accelerates diffusion models without retraining, but it inevitably introduces additional quantization noise, resulting in mean and variance deviations. In this work, we propose D2-DPM, a dual denoising mechanism aimed at precisely mitigating the adverse effects of quantization noise on the noise estimation network. Specifically, we first unravel the impact of quantization noise on the sampling equation into two components: the mean deviation and the variance deviation. The mean deviation alters the drift coefficient of the sampling equation, influencing the trajectory trend, while the variance deviation magnifies the diffusion coefficient, impacting the convergence of the sampling trajectory. The proposed D2-DPM is thus devised to denoise the quantization noise at each time step, and then denoise the noisy sample through the inverse diffusion iterations. Experimental results demonstrate that D2-DPM achieves superior generation quality, yielding a 1.42 lower FID than the full-precision model while achieving 3.99x compression and 11.67x bit-operation acceleration.
comment: 9 pages, 4 figures, acceptted by AAAI2025
☆ Object-Centric 2D Gaussian Splatting: Background Removal and Occlusion-Aware Pruning for Compact Object Models ICPR
Current Gaussian Splatting approaches are effective for reconstructing entire scenes but lack the option to target specific objects, making them computationally expensive and unsuitable for object-specific applications. We propose a novel approach that leverages object masks to enable targeted reconstruction, resulting in object-centric models. Additionally, we introduce an occlusion-aware pruning strategy to minimize the number of Gaussians without compromising quality. Our method reconstructs compact object models, yielding object-centric Gaussian and mesh representations that are up to 96\% smaller and up to 71\% faster to train compared to the baseline while retaining competitive quality. These representations are immediately usable for downstream applications such as appearance editing and physics simulation without additional processing.
comment: Accepted at ICPRAM 2025 (https://icpram.scitevents.org/Home.aspx)
☆ Benchmarking Multimodal Models for Fine-Grained Image Analysis: A Comparative Study Across Diverse Visual Features
This article introduces a benchmark designed to evaluate the capabilities of multimodal models in analyzing and interpreting images. The benchmark focuses on seven key visual aspects: main object, additional objects, background, detail, dominant colors, style, and viewpoint. A dataset of 14,580 images, generated from diverse text prompts, was used to assess the performance of seven leading multimodal models. These models were evaluated on their ability to accurately identify and describe each visual aspect, providing insights into their strengths and weaknesses for comprehensive image understanding. The findings of this benchmark have significant implications for the development and selection of multimodal models for various image analysis tasks.
comment: 6 pages, 2 tables, 2 charts
☆ Revolutionizing Communication with Deep Learning and XAI for Enhanced Arabic Sign Language Recognition
This study introduces an integrated approach to recognizing Arabic Sign Language (ArSL) using state-of-the-art deep learning models such as MobileNetV3, ResNet50, and EfficientNet-B2. These models are further enhanced by explainable AI (XAI) techniques to boost interpretability. The ArSL2018 and RGB Arabic Alphabets Sign Language (AASL) datasets are employed, with EfficientNet-B2 achieving peak accuracies of 99.48\% and 98.99\%, respectively. Key innovations include sophisticated data augmentation methods to mitigate class imbalance, implementation of stratified 5-fold cross-validation for better generalization, and the use of Grad-CAM for clear model decision transparency. The proposed system not only sets new benchmarks in recognition accuracy but also emphasizes interpretability, making it suitable for applications in healthcare, education, and inclusive communication technologies.
comment: 13 pages, 25 figures, 16 tables
☆ DM-Mamba: Dual-domain Multi-scale Mamba for MRI reconstruction
The accelerated MRI reconstruction poses a challenging ill-posed inverse problem due to the significant undersampling in k-space. Deep neural networks, such as CNNs and ViT, have shown substantial performance improvements for this task while encountering the dilemma between global receptive fields and efficient computation. To this end, this paper pioneers exploring Mamba, a new paradigm for long-range dependency modeling with linear complexity, for efficient and effective MRI reconstruction. However, directly applying Mamba to MRI reconstruction faces three significant issues: (1) Mamba's row-wise and column-wise scanning disrupts k-space's unique spectrum, leaving its potential in k-space learning unexplored. (2) Existing Mamba methods unfold feature maps with multiple lengthy scanning paths, leading to long-range forgetting and high computational burden. (3) Mamba struggles with spatially-varying contents, resulting in limited diversity of local representations. To address these, we propose a dual-domain multi-scale Mamba for MRI reconstruction from the following perspectives: (1) We pioneer vision Mamba in k-space learning. A circular scanning is customized for spectrum unfolding, benefiting the global modeling of k-space. (2) We propose a multi-scale Mamba with an efficient scanning strategy in both image and k-space domains. It mitigates long-range forgetting and achieves a better trade-off between efficiency and performance. (3) We develop a local diversity enhancement module to improve the spatially-varying representation of Mamba. Extensive experiments are conducted on three public datasets for MRI reconstruction under various undersampling patterns. Comprehensive results demonstrate that our method significantly outperforms state-of-the-art methods with lower computational cost. Implementation code will be available at https://github.com/XiaoMengLiLiLi/DM-Mamba.
☆ Energy Backdoor Attack to Deep Neural Networks
The rise of deep learning (DL) has increased computing complexity and energy use, prompting the adoption of application specific integrated circuits (ASICs) for energy-efficient edge and mobile deployment. However, recent studies have demonstrated the vulnerability of these accelerators to energy attacks. Despite the development of various inference time energy attacks in prior research, backdoor energy attacks remain unexplored. In this paper, we design an innovative energy backdoor attack against deep neural networks (DNNs) operating on sparsity-based accelerators. Our attack is carried out in two distinct phases: backdoor injection and backdoor stealthiness. Experimental results using ResNet-18 and MobileNet-V2 models trained on CIFAR-10 and Tiny ImageNet datasets show the effectiveness of our proposed attack in increasing energy consumption on trigger samples while preserving the model's performance for clean/regular inputs. This demonstrates the vulnerability of DNNs to energy backdoor attacks. The source code of our attack is available at: https://github.com/hbrachemi/energy_backdoor.
☆ Bootstrapping Corner Cases: High-Resolution Inpainting for Safety Critical Detect and Avoid for Automated Flying
Modern machine learning techniques have shown tremendous potential, especially for object detection on camera images. For this reason, they are also used to enable safety-critical automated processes such as autonomous drone flights. We present a study on object detection for Detect and Avoid, a safety critical function for drones that detects air traffic during automated flights for safety reasons. An ill-posed problem is the generation of good and especially large data sets, since detection itself is the corner case. Most models suffer from limited ground truth in raw data, \eg recorded air traffic or frontal flight with a small aircraft. It often leads to poor and critical detection rates. We overcome this problem by using inpainting methods to bootstrap the dataset such that it explicitly contains the corner cases of the raw data. We provide an overview of inpainting methods and generative models and present an example pipeline given a small annotated dataset. We validate our method by generating a high-resolution dataset, which we make publicly available and present it to an independent object detector that was fully trained on real data.
☆ Audio-visual Deepfake Detection With Local Temporal Inconsistencies ICASSP 2025
This paper proposes an audio-visual deepfake detection approach that aims to capture fine-grained temporal inconsistencies between audio and visual modalities. To achieve this, both architectural and data synthesis strategies are introduced. From an architectural perspective, a temporal distance map, coupled with an attention mechanism, is designed to capture these inconsistencies while minimizing the impact of irrelevant temporal subsequences. Moreover, we explore novel pseudo-fake generation techniques to synthesize local inconsistencies. Our approach is evaluated against state-of-the-art methods using the DFDC and FakeAVCeleb datasets, demonstrating its effectiveness in detecting audio-visual deepfakes.
comment: Accepted in ICASSP 2025
☆ SAR Strikes Back: A New Hope for RSVQA
Remote sensing visual question answering (RSVQA) is a task that automatically extracts information from satellite images and processes a question to predict the answer from the images in textual form, helping with the interpretation of the image. While different methods have been proposed to extract information from optical images with different spectral bands and resolutions, no method has been proposed to answer questions from Synthetic Aperture Radar (SAR) images. SAR images capture electromagnetic information from the scene, and are less affected by atmospheric conditions, such as clouds. In this work, our objective is to introduce SAR in the RSVQA task, finding the best way to use this modality. In our research, we carry out a study on different pipelines for the task of RSVQA taking into account information from both SAR and optical data. To this purpose, we also present a dataset that allows for the introduction of SAR images in the RSVQA framework. We propose two different models to include the SAR modality. The first one is an end-to-end method in which we add an additional encoder for the SAR modality. In the second approach, we build on a two-stage framework. First, relevant information is extracted from SAR and, optionally, optical data. This information is then translated into natural language to be used in the second step which only relies on a language model to provide the answer. We find that the second pipeline allows us to obtain good results with SAR images alone. We then try various types of fusion methods to use SAR and optical images together, finding that a fusion at the decision level achieves the best results on the proposed dataset. We show that SAR data offers additional information when fused with the optical modality, particularly for questions related to specific land cover classes, such as water areas.
comment: 26 pages, 6 figures
☆ Revisiting Birds Eye View Perception Models with Frozen Foundation Models: DINOv2 and Metric3Dv2
Birds Eye View perception models require extensive data to perform and generalize effectively. While traditional datasets often provide abundant driving scenes from diverse locations, this is not always the case. It is crucial to maximize the utility of the available training data. With the advent of large foundation models such as DINOv2 and Metric3Dv2, a pertinent question arises: can these models be integrated into existing model architectures to not only reduce the required training data but surpass the performance of current models? We choose two model architectures in the vehicle segmentation domain to alter: Lift-Splat-Shoot, and Simple-BEV. For Lift-Splat-Shoot, we explore the implementation of frozen DINOv2 for feature extraction and Metric3Dv2 for depth estimation, where we greatly exceed the baseline results by 7.4 IoU while utilizing only half the training data and iterations. Furthermore, we introduce an innovative application of Metric3Dv2's depth information as a PseudoLiDAR point cloud incorporated into the Simple-BEV architecture, replacing traditional LiDAR. This integration results in a +3 IoU improvement compared to the Camera-only model.
comment: Accepted for publication at the Electronic Imaging - Autonomous Vehicles and Machines Connference 2025
☆ RoHan: Robust Hand Detection in Operation Room
Hand-specific localization has garnered significant interest within the computer vision community. Although there are numerous datasets with hand annotations from various angles and settings, domain transfer techniques frequently struggle in surgical environments. This is mainly due to the limited availability of gloved hand instances and the unique challenges of operating rooms (ORs). Thus, hand-detection models tailored to OR settings require extensive training and expensive annotation processes. To overcome these challenges, we present "RoHan" - a novel approach for robust hand detection in the OR, leveraging advanced semi-supervised domain adaptation techniques to tackle the challenges of varying recording conditions, diverse glove colors, and occlusions common in surgical settings. Our methodology encompasses two main stages: (1) data augmentation strategy that utilizes "Artificial Gloves," a method for augmenting publicly available hand datasets with synthetic images of hands-wearing gloves; (2) semi-supervised domain adaptation pipeline that improves detection performance in real-world OR settings through iterative prediction refinement and efficient frame filtering. We evaluate our method using two datasets: simulated enterotomy repair and saphenous vein graft harvesting. "RoHan" substantially reduces the need for extensive labeling and model training, paving the way for the practical implementation of hand detection technologies in medical settings.
comment: 12 pages
☆ Change Captioning in Remote Sensing: Evolution to SAT-Cap -- A Single-Stage Transformer Approach
Change captioning has become essential for accurately describing changes in multi-temporal remote sensing data, providing an intuitive way to monitor Earth's dynamics through natural language. However, existing change captioning methods face two key challenges: high computational demands due to multistage fusion strategy, and insufficient detail in object descriptions due to limited semantic extraction from individual images. To solve these challenges, we propose SAT-Cap based on the transformers model with a single-stage feature fusion for remote sensing change captioning. In particular, SAT-Cap integrates a Spatial-Channel Attention Encoder, a Difference-Guided Fusion module, and a Caption Decoder. Compared to typical models that require multi-stage fusion in transformer encoder and fusion module, SAT-Cap uses only a simple cosine similarity-based fusion module for information integration, reducing the complexity of the model architecture. By jointly modeling spatial and channel information in Spatial-Channel Attention Encoder, our approach significantly enhances the model's ability to extract semantic information from objects in multi-temporal remote sensing images. Extensive experiments validate the effectiveness of SAT-Cap, achieving CIDEr scores of 140.23% on the LEVIR-CC dataset and 97.74% on the DUBAI-CC dataset, surpassing current state-of-the-art methods. The code and pre-trained models will be available online.
EarthView: A Large Scale Remote Sensing Dataset for Self-Supervision
This paper presents EarthView, a comprehensive dataset specifically designed for self-supervision on remote sensing data, intended to enhance deep learning applications on Earth monitoring tasks. The dataset spans 15 tera pixels of global remote-sensing data, combining imagery from a diverse range of sources, including NEON, Sentinel, and a novel release of 1m spatial resolution data from Satellogic. Our dataset provides a wide spectrum of image data with varying resolutions, harnessed from different sensors and organized coherently into an accessible HuggingFace dataset in parquet format. This data spans five years, from 2017 to 2022. Accompanying the dataset, we introduce EarthMAE, a tailored Masked Autoencoder, developed to tackle the distinct challenges of remote sensing data. Trained in a self-supervised fashion, EarthMAE effectively processes different data modalities such as hyperspectral, multispectral, topographical data, segmentation maps, and temporal structure. This model helps us show that pre-training on Satellogic data improves performance on downstream tasks. While there is still a gap to fill in MAE for heterogeneous data, we regard this innovative combination of an expansive, diverse dataset and a versatile model adapted for self-supervised learning as a stride forward in deep learning for Earth monitoring.
comment: 2nd Workshop on Computer Vision for Earth Observation (CV4EO) Applications
☆ Guiding the classification of hepatocellular carcinoma on 3D CT-scans using deep and handcrafted radiological features
Hepatocellular carcinoma is the most spread primary liver cancer across the world ($\sim$80\% of the liver tumors). The gold standard for HCC diagnosis is liver biopsy. However, in the clinical routine, expert radiologists provide a visual diagnosis by interpreting hepatic CT-scans according to a standardized protocol, the LI-RADS, which uses five radiological criteria with an associated decision tree. In this paper, we propose an automatic approach to predict histology-proven HCC from CT images in order to reduce radiologists' inter-variability. We first show that standard deep learning methods fail to accurately predict HCC from CT-scans on a challenging database, and propose a two-step approach inspired by the LI-RADS system to improve the performance. We achieve improvements from 6 to 18 points of AUC with respect to deep learning baselines trained with different architectures. We also provide clinical validation of our method, achieving results that outperform non-expert radiologists and are on par with expert ones.
comment: IEEE ISBI 2025
☆ CellOMaps: A Compact Representation for Robust Classification of Lung Adenocarcinoma Growth Patterns
Lung adenocarcinoma (LUAD) is a morphologically heterogeneous disease, characterized by five primary histological growth patterns. The classification of such patterns is crucial due to their direct relation to prognosis but the high subjectivity and observer variability pose a major challenge. Although several studies have developed machine learning methods for growth pattern classification, they either only report the predominant pattern per slide or lack proper evaluation. We propose a generalizable machine learning pipeline capable of classifying lung tissue into one of the five patterns or as non-tumor. The proposed pipeline's strength lies in a novel compact Cell Organization Maps (cellOMaps) representation that captures the cellular spatial patterns from Hematoxylin and Eosin whole slide images (WSIs). The proposed pipeline provides state-of-the-art performance on LUAD growth pattern classification when evaluated on both internal unseen slides and external datasets, significantly outperforming the current approaches. In addition, our preliminary results show that the model's outputs can be used to predict patients Tumor Mutational Burden (TMB) levels.
☆ AgentPose: Progressive Distribution Alignment via Feature Agent for Human Pose Distillation
Pose distillation is widely adopted to reduce model size in human pose estimation. However, existing methods primarily emphasize the transfer of teacher knowledge while often neglecting the performance degradation resulted from the curse of capacity gap between teacher and student. To address this issue, we propose AgentPose, a novel pose distillation method that integrates a feature agent to model the distribution of teacher features and progressively aligns the distribution of student features with that of the teacher feature, effectively overcoming the capacity gap and enhancing the ability of knowledge transfer. Our comprehensive experiments conducted on the COCO dataset substantiate the effectiveness of our method in knowledge transfer, particularly in scenarios with a high capacity gap.
comment: 5 pages, 1 figures
☆ Benchmarking Vision Foundation Models for Input Monitoring in Autonomous Driving
Deep neural networks (DNNs) remain challenged by distribution shifts in complex open-world domains like automated driving (AD): Absolute robustness against yet unknown novel objects (semantic shift) or styles like lighting conditions (covariate shift) cannot be guaranteed. Hence, reliable operation-time monitors for identification of out-of-training-data-distribution (OOD) scenarios are imperative. Current approaches for OOD classification are untested for complex domains like AD, are limited in the kinds of shifts they detect, or even require supervision with OOD samples. To prepare for unanticipated shifts, we instead establish a framework around a principled, unsupervised, and model-agnostic method that unifies detection of all kinds of shifts: Find a full model of the training data's feature distribution, to then use its density at new points as in-distribution (ID) score. To implement this, we propose to combine the newly available Vision Foundation Models (VFM) as feature extractors with one of four alternative density modeling techniques. In an extensive benchmark of 4 VFMs against 20 baselines, we show the superior performance of VFM feature encodings compared to shift-specific OOD monitors. Additionally, we find that sophisticated architectures outperform larger latent space dimensionality; and our method identifies samples with higher risk of errors on downstream tasks, despite being model-agnostic. This suggests that VFMs are promising to realize model-agnostic, unsupervised, reliable safety monitors in complex vision tasks.
☆ Skeleton and Font Generation Network for Zero-shot Chinese Character Generation
Automatic font generation remains a challenging research issue, primarily due to the vast number of Chinese characters, each with unique and intricate structures. Our investigation of previous studies reveals inherent bias capable of causing structural changes in characters. Specifically, when generating a Chinese character similar to, but different from, those in the training samples, the bias is prone to either correcting or ignoring these subtle variations. To address this concern, we propose a novel Skeleton and Font Generation Network (SFGN) to achieve a more robust Chinese character font generation. Our approach includes a skeleton builder and font generator. The skeleton builder synthesizes content features using low-resource text input, enabling our technique to realize font generation independently of content image inputs. Unlike previous font generation methods that treat font style as a global embedding, we introduce a font generator to align content and style features on the radical level, which is a brand-new perspective for font generation. Except for common characters, we also conduct experiments on misspelled characters, a substantial portion of which slightly differs from the common ones. Our approach visually demonstrates the efficacy of generated images and outperforms current state-of-the-art font generation methods. Moreover, we believe that misspelled character generation have significant pedagogical implications and verify such supposition through experiments. We used generated misspelled characters as data augmentation in Chinese character error correction tasks, simulating the scenario where students learn handwritten Chinese characters with the help of misspelled characters. The significantly improved performance of error correction tasks demonstrates the effectiveness of our proposed approach and the value of misspelled character generation.
comment: 36 pages, 10 figures
☆ Self-Attentive Spatio-Temporal Calibration for Precise Intermediate Layer Matching in ANN-to-SNN Distillation
Spiking Neural Networks (SNNs) are promising for low-power computation due to their event-driven mechanism but often suffer from lower accuracy compared to Artificial Neural Networks (ANNs). ANN-to-SNN knowledge distillation can improve SNN performance, but previous methods either focus solely on label information, missing valuable intermediate layer features, or use a layer-wise approach that neglects spatial and temporal semantic inconsistencies, leading to performance degradation.To address these limitations, we propose a novel method called self-attentive spatio-temporal calibration (SASTC). SASTC uses self-attention to identify semantically aligned layer pairs between ANN and SNN, both spatially and temporally. This enables the autonomous transfer of relevant semantic information. Extensive experiments show that SASTC outperforms existing methods, effectively solving the mismatching problem. Superior accuracy results include 95.12% on CIFAR-10, 79.40% on CIFAR-100 with 2 time steps, and 68.69% on ImageNet with 4 time steps for static datasets, and 97.92% on DVS-Gesture and 83.60% on DVS-CIFAR10 for neuromorphic datasets. This marks the first time SNNs have outperformed ANNs on both CIFAR-10 and CIFAR-100, shedding the new light on the potential applications of SNNs.
☆ Exploring visual language models as a powerful tool in the diagnosis of Ewing Sarcoma
Ewing's sarcoma (ES), characterized by a high density of small round blue cells without structural organization, presents a significant health concern, particularly among adolescents aged 10 to 19. Artificial intelligence-based systems for automated analysis of histopathological images are promising to contribute to an accurate diagnosis of ES. In this context, this study explores the feature extraction ability of different pre-training strategies for distinguishing ES from other soft tissue or bone sarcomas with similar morphology in digitized tissue microarrays for the first time, as far as we know. Vision-language supervision (VLS) is compared to fully-supervised ImageNet pre-training within a multiple instance learning paradigm. Our findings indicate a substantial improvement in diagnostic accuracy with the adaption of VLS using an in-domain dataset. Notably, these models not only enhance the accuracy of predicted classes but also drastically reduce the number of trainable parameters and computational costs.
comment: 11 pages, 5 figures, 2 tables. Oral presentation at KES-InMed 2024 held in Madeira, Portugal
☆ Robust Low-Light Human Pose Estimation through Illumination-Texture Modulation
As critical visual details become obscured, the low visibility and high ISO noise in extremely low-light images pose a significant challenge to human pose estimation. Current methods fail to provide high-quality representations due to reliance on pixel-level enhancements that compromise semantics and the inability to effectively handle extreme low-light conditions for robust feature learning. In this work, we propose a frequency-based framework for low-light human pose estimation, rooted in the "divide-and-conquer" principle. Instead of uniformly enhancing the entire image, our method focuses on task-relevant information. By applying dynamic illumination correction to the low-frequency components and low-rank denoising to the high-frequency components, we effectively enhance both the semantic and texture information essential for accurate pose estimation. As a result, this targeted enhancement method results in robust, high-quality representations, significantly improving pose estimation performance. Extensive experiments demonstrating its superiority over state-of-the-art methods in various challenging low-light scenarios.
comment: 5 pages, 2 figures, conference
☆ DisCoPatch: Batch Statistics Are All You Need For OOD Detection, But Only If You Can Trust Them
Out-of-distribution (OOD) detection holds significant importance across many applications. While semantic and domain-shift OOD problems are well-studied, this work focuses on covariate shifts - subtle variations in the data distribution that can degrade machine learning performance. We hypothesize that detecting these subtle shifts can improve our understanding of in-distribution boundaries, ultimately improving OOD detection. In adversarial discriminators trained with Batch Normalization (BN), real and adversarial samples form distinct domains with unique batch statistics - a property we exploit for OOD detection. We introduce DisCoPatch, an unsupervised Adversarial Variational Autoencoder (VAE) framework that harnesses this mechanism. During inference, batches consist of patches from the same image, ensuring a consistent data distribution that allows the model to rely on batch statistics. DisCoPatch uses the VAE's suboptimal outputs (generated and reconstructed) as negative samples to train the discriminator, thereby improving its ability to delineate the boundary between in-distribution samples and covariate shifts. By tightening this boundary, DisCoPatch achieves state-of-the-art results in public OOD detection benchmarks. The proposed model not only excels in detecting covariate shifts, achieving 95.5% AUROC on ImageNet-1K(-C) but also outperforms all prior methods on public Near-OOD (95.0%) benchmarks. With a compact model size of 25MB, it achieves high OOD detection performance at notably lower latency than existing methods, making it an efficient and practical solution for real-world OOD detection applications. The code will be made publicly available
☆ Maximizing Uncertainty for Federated learning via Bayesian Optimisation-based Model Poisoning
As we transition from Narrow Artificial Intelligence towards Artificial Super Intelligence, users are increasingly concerned about their privacy and the trustworthiness of machine learning (ML) technology. A common denominator for the metrics of trustworthiness is the quantification of uncertainty inherent in DL algorithms, and specifically in the model parameters, input data, and model predictions. One of the common approaches to address privacy-related issues in DL is to adopt distributed learning such as federated learning (FL), where private raw data is not shared among users. Despite the privacy-preserving mechanisms in FL, it still faces challenges in trustworthiness. Specifically, the malicious users, during training, can systematically create malicious model parameters to compromise the models predictive and generative capabilities, resulting in high uncertainty about their reliability. To demonstrate malicious behaviour, we propose a novel model poisoning attack method named Delphi which aims to maximise the uncertainty of the global model output. We achieve this by taking advantage of the relationship between the uncertainty and the model parameters of the first hidden layer of the local model. Delphi employs two types of optimisation , Bayesian Optimisation and Least Squares Trust Region, to search for the optimal poisoned model parameters, named as Delphi-BO and Delphi-LSTR. We quantify the uncertainty using the KL Divergence to minimise the distance of the predictive probability distribution towards an uncertain distribution of model output. Furthermore, we establish a mathematical proof for the attack effectiveness demonstrated in FL. Numerical results demonstrate that Delphi-BO induces a higher amount of uncertainty than Delphi-LSTR highlighting vulnerability of FL systems to model poisoning attacks.
comment: 14 pages
☆ Combining imaging and shape features for prediction tasks of Alzheimer's disease classification and brain age regression
We investigate combining imaging and shape features extracted from MRI for the clinically relevant tasks of brain age prediction and Alzheimer's disease classification. Our proposed model fuses ResNet-extracted image embeddings with shape embeddings from a bespoke graph neural network. The shape embeddings are derived from surface meshes of 15 brain structures, capturing detailed geometric information. Combined with the appearance features from T1-weighted images, we observe improvements in the prediction performance on both tasks, with substantial gains for classification. We evaluate the model using public datasets, including CamCAN, IXI, and OASIS3, demonstrating the effectiveness of fusing imaging and shape features for brain analysis.
☆ GAC-Net_Geometric and attention-based Network for Depth Completion
Depth completion is a key task in autonomous driving, aiming to complete sparse LiDAR depth measurements into high-quality dense depth maps through image guidance. However, existing methods usually treat depth maps as an additional channel of color images, or directly perform convolution on sparse data, failing to fully exploit the 3D geometric information in depth maps, especially with limited performance in complex boundaries and sparse areas. To address these issues, this paper proposes a depth completion network combining channel attention mechanism and 3D global feature perception (CGA-Net). The main innovations include: 1) Utilizing PointNet++ to extract global 3D geometric features from sparse depth maps, enhancing the scene perception ability of low-line LiDAR data; 2) Designing a channel-attention-based multimodal feature fusion module to efficiently integrate sparse depth, RGB images, and 3D geometric features; 3) Combining residual learning with CSPN++ to optimize the depth refinement stage, further improving the completion quality in edge areas and complex scenes. Experiments on the KITTI depth completion dataset show that CGA-Net can significantly improve the prediction accuracy of dense depth maps, achieving a new state-of-the-art (SOTA), and demonstrating strong robustness to sparse and complex scenes.
comment: 13pages,4 figures, 2 tables
☆ Threshold Attention Network for Semantic Segmentation of Remote Sensing Images
Semantic segmentation of remote sensing images is essential for various applications, including vegetation monitoring, disaster management, and urban planning. Previous studies have demonstrated that the self-attention mechanism (SA) is an effective approach for designing segmentation networks that can capture long-range pixel dependencies. SA enables the network to model the global dependencies between the input features, resulting in improved segmentation outcomes. However, the high density of attentional feature maps used in this mechanism causes exponential increases in computational complexity. Additionally, it introduces redundant information that negatively impacts the feature representation. Inspired by traditional threshold segmentation algorithms, we propose a novel threshold attention mechanism (TAM). This mechanism significantly reduces computational effort while also better modeling the correlation between different regions of the feature map. Based on TAM, we present a threshold attention network (TANet) for semantic segmentation. TANet consists of an attentional feature enhancement module (AFEM) for global feature enhancement of shallow features and a threshold attention pyramid pooling module (TAPP) for acquiring feature information at different scales for deep features. We have conducted extensive experiments on the ISPRS Vaihingen and Potsdam datasets. The results demonstrate the validity and superiority of our proposed TANet compared to the most state-of-the-art models.
☆ V-Trans4Style: Visual Transition Recommendation for Video Production Style Adaptation
We introduce V-Trans4Style, an innovative algorithm tailored for dynamic video content editing needs. It is designed to adapt videos to different production styles like documentaries, dramas, feature films, or a specific YouTube channel's video-making technique. Our algorithm recommends optimal visual transitions to help achieve this flexibility using a more bottom-up approach. We first employ a transformer-based encoder-decoder network to learn recommending temporally consistent and visually seamless sequences of visual transitions using only the input videos. We then introduce a style conditioning module that leverages this model to iteratively adjust the visual transitions obtained from the decoder through activation maximization. We demonstrate the efficacy of our method through experiments conducted on our newly introduced AutoTransition++ dataset. It is a 6k video version of AutoTransition Dataset that additionally categorizes its videos into different production style categories. Our encoder-decoder model outperforms the state-of-the-art transition recommendation method, achieving improvements of 10% to 80% in Recall@K and mean rank values over baseline. Our style conditioning module results in visual transitions that improve the capture of the desired video production style characteristics by an average of around 12% in comparison to other methods when measured with similarity metrics. We hope that our work serves as a foundation for exploring and understanding video production styles further.
☆ Facial Dynamics in Video: Instruction Tuning for Improved Facial Expression Perception and Contextual Awareness
Facial expression captioning has found widespread application across various domains. Recently, the emergence of video Multimodal Large Language Models (MLLMs) has shown promise in general video understanding tasks. However, describing facial expressions within videos poses two major challenges for these models: (1) the lack of adequate datasets and benchmarks, and (2) the limited visual token capacity of video MLLMs. To address these issues, this paper introduces a new instruction-following dataset tailored for dynamic facial expression caption. The dataset comprises 5,033 high-quality video clips annotated manually, containing over 700,000 tokens. Its purpose is to improve the capability of video MLLMs to discern subtle facial nuances. Furthermore, we propose FaceTrack-MM, which leverages a limited number of tokens to encode the main character's face. This model demonstrates superior performance in tracking faces and focusing on the facial expressions of the main characters, even in intricate multi-person scenarios. Additionally, we introduce a novel evaluation metric combining event extraction, relation classification, and the longest common subsequence (LCS) algorithm to assess the content consistency and temporal sequence consistency of generated text. Moreover, we present FEC-Bench, a benchmark designed to assess the performance of existing video MLLMs in this specific task. All data and source code will be made publicly available.
☆ Zero-shot Video Moment Retrieval via Off-the-shelf Multimodal Large Language Models AAAI 2025
The target of video moment retrieval (VMR) is predicting temporal spans within a video that semantically match a given linguistic query. Existing VMR methods based on multimodal large language models (MLLMs) overly rely on expensive high-quality datasets and time-consuming fine-tuning. Although some recent studies introduce a zero-shot setting to avoid fine-tuning, they overlook inherent language bias in the query, leading to erroneous localization. To tackle the aforementioned challenges, this paper proposes Moment-GPT, a tuning-free pipeline for zero-shot VMR utilizing frozen MLLMs. Specifically, we first employ LLaMA-3 to correct and rephrase the query to mitigate language bias. Subsequently, we design a span generator combined with MiniGPT-v2 to produce candidate spans adaptively. Finally, to leverage the video comprehension capabilities of MLLMs, we apply VideoChatGPT and span scorer to select the most appropriate spans. Our proposed method substantially outperforms the state-ofthe-art MLLM-based and zero-shot models on several public datasets, including QVHighlights, ActivityNet-Captions, and Charades-STA.
comment: Accepted by AAAI 2025
☆ SkipClick: Combining Quick Responses and Low-Level Features for Interactive Segmentation in Winter Sports Contexts
In this paper, we present a novel architecture for interactive segmentation in winter sports contexts. The field of interactive segmentation deals with the prediction of high-quality segmentation masks by informing the network about the objects position with the help of user guidance. In our case the guidance consists of click prompts. For this task, we first present a baseline architecture which is specifically geared towards quickly responding after each click. Afterwards, we motivate and describe a number of architectural modifications which improve the performance when tasked with segmenting winter sports equipment on the WSESeg dataset. With regards to the average NoC@85 metric on the WSESeg classes, we outperform SAM and HQ-SAM by 2.336 and 7.946 clicks, respectively. When applied to the HQSeg-44k dataset, our system delivers state-of-the-art results with a NoC@90 of 6.00 and NoC@95 of 9.89. In addition to that, we test our model on a novel dataset containing masks for humans during skiing.
comment: 4 figures, 6 tables, 12 pages
☆ AI Guide Dog: Egocentric Path Prediction on Smartphone
This paper introduces AI Guide Dog (AIGD), a lightweight egocentric navigation assistance system for visually impaired individuals, designed for real-time deployment on smartphones. AIGD addresses key challenges in blind navigation by employing a vision-only, multi-label classification approach to predict directional commands, ensuring safe traversal across diverse environments. We propose a novel technique to enable goal-based outdoor navigation by integrating GPS signals and high-level directions, while also addressing uncertain multi-path predictions for destination-free indoor navigation. Our generalized model is the first navigation assistance system to handle both goal-oriented and exploratory navigation scenarios across indoor and outdoor settings, establishing a new state-of-the-art in blind navigation. We present methods, datasets, evaluations, and deployment insights to encourage further innovations in assistive navigation systems.
☆ Robust Hyperspectral Image Panshapring via Sparse Spatial-Spectral Representation
High-resolution hyperspectral imaging plays a crucial role in various remote sensing applications, yet its acquisition often faces fundamental limitations due to hardware constraints. This paper introduces S$^{3}$RNet, a novel framework for hyperspectral image pansharpening that effectively combines low-resolution hyperspectral images (LRHSI) with high-resolution multispectral images (HRMSI) through sparse spatial-spectral representation. The core of S$^{3}$RNet is the Multi-Branch Fusion Network (MBFN), which employs parallel branches to capture complementary features at different spatial and spectral scales. Unlike traditional approaches that treat all features equally, our Spatial-Spectral Attention Weight Block (SSAWB) dynamically adjusts feature weights to maintain sparse representation while suppressing noise and redundancy. To enhance feature propagation, we incorporate the Dense Feature Aggregation Block (DFAB), which efficiently aggregates inputted features through dense connectivity patterns. This integrated design enables S$^{3}$RNet to selectively emphasize the most informative features from differnt scale while maintaining computational efficiency. Comprehensive experiments demonstrate that S$^{3}$RNet achieves state-of-the-art performance across multiple evaluation metrics, showing particular strength in maintaining high reconstruction quality even under challenging noise conditions. The code will be made publicly available.
comment: Submitted to IGARSS 2025
☆ Early prediction of the transferability of bovine embryos from videomicroscopy
Videomicroscopy is a promising tool combined with machine learning for studying the early development of in vitro fertilized bovine embryos and assessing its transferability as soon as possible. We aim to predict the embryo transferability within four days at most, taking 2D time-lapse microscopy videos as input. We formulate this problem as a supervised binary classification problem for the classes transferable and not transferable. The challenges are three-fold: 1) poorly discriminating appearance and motion, 2) class ambiguity, 3) small amount of annotated data. We propose a 3D convolutional neural network involving three pathways, which makes it multi-scale in time and able to handle appearance and motion in different ways. For training, we retain the focal loss. Our model, named SFR, compares favorably to other methods. Experiments demonstrate its effectiveness and accuracy for our challenging biological task.
comment: Accepted at the 2024 IEEE International Conference on Image Processing
☆ VENOM: Text-driven Unrestricted Adversarial Example Generation with Diffusion Models
Adversarial attacks have proven effective in deceiving machine learning models by subtly altering input images, motivating extensive research in recent years. Traditional methods constrain perturbations within $l_p$-norm bounds, but advancements in Unrestricted Adversarial Examples (UAEs) allow for more complex, generative-model-based manipulations. Diffusion models now lead UAE generation due to superior stability and image quality over GANs. However, existing diffusion-based UAE methods are limited to using reference images and face challenges in generating Natural Adversarial Examples (NAEs) directly from random noise, often producing uncontrolled or distorted outputs. In this work, we introduce VENOM, the first text-driven framework for high-quality unrestricted adversarial examples generation through diffusion models. VENOM unifies image content generation and adversarial synthesis into a single reverse diffusion process, enabling high-fidelity adversarial examples without sacrificing attack success rate (ASR). To stabilize this process, we incorporate an adaptive adversarial guidance strategy with momentum, ensuring that the generated adversarial examples $x^*$ align with the distribution $p(x)$ of natural images. Extensive experiments demonstrate that VENOM achieves superior ASR and image quality compared to prior methods, marking a significant advancement in adversarial example generation and providing insights into model vulnerabilities for improved defense development.
☆ Cloud Removal With PolSAR-Optical Data Fusion Using A Two-Flow Residual Network
Optical remote sensing images play a crucial role in the observation of the Earth's surface. However, obtaining complete optical remote sensing images is challenging due to cloud cover. Reconstructing cloud-free optical images has become a major task in recent years. This paper presents a two-flow Polarimetric Synthetic Aperture Radar (PolSAR)-Optical data fusion cloud removal algorithm (PODF-CR), which achieves the reconstruction of missing optical images. PODF-CR consists of an encoding module and a decoding module. The encoding module includes two parallel branches that extract PolSAR image features and optical image features. To address speckle noise in PolSAR images, we introduce dynamic filters in the PolSAR branch for image denoising. To better facilitate the fusion between multimodal optical images and PolSAR images, we propose fusion blocks based on cross-skip connections to enable interaction of multimodal data information. The obtained fusion features are refined through an attention mechanism to provide better conditions for the subsequent decoding of the fused images. In the decoding module, multi-scale convolution is introduced to obtain multi-scale information. Additionally, to better utilize comprehensive scattering information and polarization characteristics to assist in the restoration of optical images, we use a dataset for cloud restoration called OPT-BCFSAR-PFSAR, which includes backscatter coefficient feature images and polarization feature images obtained from PoLSAR data and optical images. Experimental results demonstrate that this method outperforms existing methods in both qualitative and quantitative evaluations.
☆ Demographic Variability in Face Image Quality Measures
Face image quality assessment (FIQA) algorithms are being integrated into online identity management applications. These applications allow users to upload a face image as part of their document issuance process, where the image is then run through a quality assessment process to make sure it meets the quality and compliance requirements. Concerns about demographic bias have been raised about biometric systems, given the societal implications this may cause. It is therefore important that demographic variability in FIQA algorithms is assessed such that mitigation measures can be created. In this work, we study the demographic variability of all face image quality measures included in the ISO/IEC 29794-5 international standard across three demographic variables: age, gender, and skin tone. The results are rather promising and show no clear bias toward any specific demographic group for most measures. Only two quality measures are found to have considerable variations in their outcomes for different groups on the skin tone variable.
☆ Tarsier2: Advancing Large Vision-Language Models from Detailed Video Description to Comprehensive Video Understanding
We introduce Tarsier2, a state-of-the-art large vision-language model (LVLM) designed for generating detailed and accurate video descriptions, while also exhibiting superior general video understanding capabilities. Tarsier2 achieves significant advancements through three key upgrades: (1) Scaling pre-training data from 11M to 40M video-text pairs, enriching both volume and diversity; (2) Performing fine-grained temporal alignment during supervised fine-tuning; (3) Using model-based sampling to automatically construct preference data and applying DPO training for optimization. Extensive experiments show that Tarsier2-7B consistently outperforms leading proprietary models, including GPT-4o and Gemini 1.5 Pro, in detailed video description tasks. On the DREAM-1K benchmark, Tarsier2-7B improves F1 by 2.8\% over GPT-4o and 5.8\% over Gemini-1.5-Pro. In human side-by-side evaluations, Tarsier2-7B shows a +8.6\% performance advantage over GPT-4o and +24.9\% over Gemini-1.5-Pro. Tarsier2-7B also sets new state-of-the-art results across 15 public benchmarks, spanning tasks such as video question-answering, video grounding, hallucination test, and embodied question-answering, demonstrating its versatility as a robust generalist vision-language model.
☆ Mitigating Algorithmic Bias in Multiclass CNN Classifications Using Causal Modeling
This study describes a procedure for applying causal modeling to detect and mitigate algorithmic bias in a multiclass classification problem. The dataset was derived from the FairFace dataset, supplemented with emotional labels generated by the DeepFace pre-trained model. A custom Convolutional Neural Network (CNN) was developed, consisting of four convolutional blocks, followed by fully connected layers and dropout layers to mitigate overfitting. Gender bias was identified in the CNN model's classifications: Females were more likely to be classified as "happy" or "sad," while males were more likely to be classified as "neutral." To address this, the one-vs-all (OvA) technique was applied. A causal model was constructed for each emotion class to adjust the CNN model's predicted class probabilities. The adjusted probabilities for the various classes were then aggregated by selecting the class with the highest probability. The resulting debiased classifications demonstrated enhanced gender fairness across all classes, with negligible impact--or even a slight improvement--on overall accuracy. This study highlights that algorithmic fairness and accuracy are not necessarily trade-offs. All data and code for this study are publicly available for download.
comment: 7 pages; 6 figures
☆ Make-A-Character 2: Animatable 3D Character Generation From a Single Image
This report introduces Make-A-Character 2, an advanced system for generating high-quality 3D characters from single portrait photographs, ideal for game development and digital human applications. Make-A-Character 2 builds upon its predecessor by incorporating several significant improvements for image-based head generation. We utilize the IC-Light method to correct non-ideal illumination in input photos and apply neural network-based color correction to harmonize skin tones between the photos and game engine renders. We also employ the Hierarchical Representation Network to capture high-frequency facial structures and conduct adaptive skeleton calibration for accurate and expressive facial animations. The entire image-to-3D-character generation process takes less than 2 minutes. Furthermore, we leverage transformer architecture to generate co-speech facial and gesture actions, enabling real-time conversation with the generated character. These technologies have been integrated into our conversational AI avatar products.
comment: Technical Report
☆ deepTerra -- AI Land Classification Made Easy
deepTerra is a comprehensive platform designed to facilitate the classification of land surface features using machine learning and satellite imagery. The platform includes modules for data collection, image augmentation, training, testing, and prediction, streamlining the entire workflow for image classification tasks. This paper presents a detailed overview of the capabilities of deepTerra, shows how it has been applied to various research areas, and discusses the future directions it might take.
☆ State-of-the-Art Transformer Models for Image Super-Resolution: Techniques, Challenges, and Applications
Image Super-Resolution (SR) aims to recover a high-resolution image from its low-resolution counterpart, which has been affected by a specific degradation process. This is achieved by enhancing detail and visual quality. Recent advancements in transformer-based methods have remolded image super-resolution by enabling high-quality reconstructions surpassing previous deep-learning approaches like CNN and GAN-based. This effectively addresses the limitations of previous methods, such as limited receptive fields, poor global context capture, and challenges in high-frequency detail recovery. Additionally, the paper reviews recent trends and advancements in transformer-based SR models, exploring various innovative techniques and architectures that combine transformers with traditional networks to balance global and local contexts. These neoteric methods are critically analyzed, revealing promising yet unexplored gaps and potential directions for future research. Several visualizations of models and techniques are included to foster a holistic understanding of recent trends. This work seeks to offer a structured roadmap for researchers at the forefront of deep learning, specifically exploring the impact of transformers on super-resolution techniques.
comment: 8 pages
☆ An Intra- and Cross-frame Topological Consistency Scheme for Semi-supervised Atherosclerotic Coronary Plaque Segmentation ICASSP 2025
Enhancing the precision of segmenting coronary atherosclerotic plaques from CT Angiography (CTA) images is pivotal for advanced Coronary Atherosclerosis Analysis (CAA), which distinctively relies on the analysis of vessel cross-section images reconstructed via Curved Planar Reformation. This task presents significant challenges due to the indistinct boundaries and structures of plaques and blood vessels, leading to the inadequate performance of current deep learning models, compounded by the inherent difficulty in annotating such complex data. To address these issues, we propose a novel dual-consistency semi-supervised framework that integrates Intra-frame Topological Consistency (ITC) and Cross-frame Topological Consistency (CTC) to leverage labeled and unlabeled data. ITC employs a dual-task network for simultaneous segmentation mask and Skeleton-aware Distance Transform (SDT) prediction, achieving similar prediction of topology structure through consistency constraint without additional annotations. Meanwhile, CTC utilizes an unsupervised estimator for analyzing pixel flow between skeletons and boundaries of adjacent frames, ensuring spatial continuity. Experiments on two CTA datasets show that our method surpasses existing semi-supervised methods and approaches the performance of supervised methods on CAA. In addition, our method also performs better than other methods on the ACDC dataset, demonstrating its generalization.
comment: Accepted by ICASSP 2025
☆ 3UR-LLM: An End-to-End Multimodal Large Language Model for 3D Scene Understanding
Multi-modal Large Language Models (MLLMs) exhibit impressive capabilities in 2D tasks, yet encounter challenges in discerning the spatial positions, interrelations, and causal logic in scenes when transitioning from 2D to 3D representations. We find that the limitations mainly lie in: i) the high annotation cost restricting the scale-up of volumes of 3D scene data, and ii) the lack of a straightforward and effective way to perceive 3D information which results in prolonged training durations and complicates the streamlined framework. To this end, we develop pipeline based on open-source 2D MLLMs and LLMs to generate high-quality 3D-text pairs and construct 3DS-160K , to enhance the pre-training process. Leveraging this high-quality pre-training data, we introduce the 3UR-LLM model, an end-to-end 3D MLLM designed for precise interpretation of 3D scenes, showcasing exceptional capability in navigating the complexities of the physical world. 3UR-LLM directly receives 3D point cloud as input and project 3D features fused with text instructions into a manageable set of tokens. Considering the computation burden derived from these hybrid tokens, we design a 3D compressor module to cohesively compress the 3D spatial cues and textual narrative. 3UR-LLM achieves promising performance with respect to the previous SOTAs, for instance, 3UR-LLM exceeds its counterparts by 7.1\% CIDEr on ScanQA, while utilizing fewer training resources. The code and model weights for 3UR-LLM and the 3DS-160K benchmark are available at 3UR-LLM.
comment: Accepted to IEEE Transactions on Multimedia (TMM)
☆ AVS-Mamba: Exploring Temporal and Multi-modal Mamba for Audio-Visual Segmentation
The essence of audio-visual segmentation (AVS) lies in locating and delineating sound-emitting objects within a video stream. While Transformer-based methods have shown promise, their handling of long-range dependencies struggles due to quadratic computational costs, presenting a bottleneck in complex scenarios. To overcome this limitation and facilitate complex multi-modal comprehension with linear complexity, we introduce AVS-Mamba, a selective state space model to address the AVS task. Our framework incorporates two key components for video understanding and cross-modal learning: Temporal Mamba Block for sequential video processing and Vision-to-Audio Fusion Block for advanced audio-vision integration. Building on this, we develop the Multi-scale Temporal Encoder, aimed at enhancing the learning of visual features across scales, facilitating the perception of intra- and inter-frame information. To perform multi-modal fusion, we propose the Modality Aggregation Decoder, leveraging the Vision-to-Audio Fusion Block to integrate visual features into audio features across both frame and temporal levels. Further, we adopt the Contextual Integration Pyramid to perform audio-to-vision spatial-temporal context collaboration. Through these innovative contributions, our approach achieves new state-of-the-art results on the AVSBench-object and AVSBench-semantic datasets. Our source code and model weights are available at AVS-Mamba.
comment: Accepted to IEEE Transactions on Multimedia (TMM)
☆ A Low-cost and Ultra-lightweight Binary Neural Network for Traffic Signal Recognition
The deployment of neural networks in vehicle platforms and wearable Artificial Intelligence-of-Things (AIOT) scenarios has become a research area that has attracted much attention. With the continuous evolution of deep learning technology, many image classification models are committed to improving recognition accuracy, but this is often accompanied by problems such as large model resource usage, complex structure, and high power consumption, which makes it challenging to deploy on resource-constrained platforms. Herein, we propose an ultra-lightweight binary neural network (BNN) model designed for hardware deployment, and conduct image classification research based on the German Traffic Sign Recognition Benchmark (GTSRB) dataset. In addition, we also verify it on the Chinese Traffic Sign (CTS) and Belgian Traffic Sign (BTS) datasets. The proposed model shows excellent recognition performance with an accuracy of up to 97.64%, making it one of the best performing BNN models in the GTSRB dataset. Compared with the full-precision model, the accuracy loss is controlled within 1%, and the parameter storage overhead of the model is only 10% of that of the full-precision model. More importantly, our network model only relies on logical operations and low-bit width fixed-point addition and subtraction operations during the inference phase, which greatly simplifies the design complexity of the processing element (PE). Our research shows the great potential of BNN in the hardware deployment of computer vision models, especially in the field of computer vision tasks related to autonomous driving.
☆ Learning Motion and Temporal Cues for Unsupervised Video Object Segmentation
In this paper, we address the challenges in unsupervised video object segmentation (UVOS) by proposing an efficient algorithm, termed MTNet, which concurrently exploits motion and temporal cues. Unlike previous methods that focus solely on integrating appearance with motion or on modeling temporal relations, our method combines both aspects by integrating them within a unified framework. MTNet is devised by effectively merging appearance and motion features during the feature extraction process within encoders, promoting a more complementary representation. To capture the intricate long-range contextual dynamics and information embedded within videos, a temporal transformer module is introduced, facilitating efficacious inter-frame interactions throughout a video clip. Furthermore, we employ a cascade of decoders all feature levels across all feature levels to optimally exploit the derived features, aiming to generate increasingly precise segmentation masks. As a result, MTNet provides a strong and compact framework that explores both temporal and cross-modality knowledge to robustly localize and track the primary object accurately in various challenging scenarios efficiently. Extensive experiments across diverse benchmarks conclusively show that our method not only attains state-of-the-art performance in unsupervised video object segmentation but also delivers competitive results in video salient object detection. These findings highlight the method's robust versatility and its adeptness in adapting to a range of segmentation tasks. Source code is available on https://github.com/hy0523/MTNet.
comment: Accepted to IEEE Transactions on Neural Networks and Learning Systems (TNNLS)
☆ Balance Divergence for Knowledge Distillation
Knowledge distillation has been widely adopted in computer vision task processing, since it can effectively enhance the performance of lightweight student networks by leveraging the knowledge transferred from cumbersome teacher networks. Most existing knowledge distillation methods utilize Kullback-Leibler divergence to mimic the logit output probabilities between the teacher network and the student network. Nonetheless, these methods may neglect the negative parts of the teacher's ''dark knowledge'' because the divergence calculations may ignore the effect of the minute probabilities from the teacher's logit output. This deficiency may lead to suboptimal performance in logit mimicry during the distillation process and result in an imbalance of information acquired by the student network. In this paper, we investigate the impact of this imbalance and propose a novel method, named Balance Divergence Distillation. By introducing a compensatory operation using reverse Kullback-Leibler divergence, our method can improve the modeling of the extremely small values in the negative from the teacher and preserve the learning capacity for the positive. Furthermore, we test the impact of different temperature coefficients adjustments, which may conducted to further balance for knowledge transferring. We evaluate the proposed method on several computer vision tasks, including image classification and semantic segmentation. The evaluation results show that our method achieves an accuracy improvement of 1%~3% for lightweight students on both CIFAR-100 and ImageNet dataset, and a 4.55% improvement in mIoU for PSP-ResNet18 on the Cityscapes dataset. The experiments show that our method is a simple yet highly effective solution that can be smoothly applied to different knowledge distillation methods.
☆ BioPose: Biomechanically-accurate 3D Pose Estimation from Monocular Videos
Recent advancements in 3D human pose estimation from single-camera images and videos have relied on parametric models, like SMPL. However, these models oversimplify anatomical structures, limiting their accuracy in capturing true joint locations and movements, which reduces their applicability in biomechanics, healthcare, and robotics. Biomechanically accurate pose estimation, on the other hand, typically requires costly marker-based motion capture systems and optimization techniques in specialized labs. To bridge this gap, we propose BioPose, a novel learning-based framework for predicting biomechanically accurate 3D human pose directly from monocular videos. BioPose includes three key components: a Multi-Query Human Mesh Recovery model (MQ-HMR), a Neural Inverse Kinematics (NeurIK) model, and a 2D-informed pose refinement technique. MQ-HMR leverages a multi-query deformable transformer to extract multi-scale fine-grained image features, enabling precise human mesh recovery. NeurIK treats the mesh vertices as virtual markers, applying a spatial-temporal network to regress biomechanically accurate 3D poses under anatomical constraints. To further improve 3D pose estimations, a 2D-informed refinement step optimizes the query tokens during inference by aligning the 3D structure with 2D pose observations. Experiments on benchmark datasets demonstrate that BioPose significantly outperforms state-of-the-art methods. Project website: \url{https://m-usamasaleem.github.io/publication/BioPose/BioPose.html}.
☆ Parameter-Inverted Image Pyramid Networks for Visual Perception and Multimodal Understanding
Image pyramids are widely adopted in top-performing methods to obtain multi-scale features for precise visual perception and understanding. However, current image pyramids use the same large-scale model to process multiple resolutions of images, leading to significant computational cost. To address this challenge, we propose a novel network architecture, called Parameter-Inverted Image Pyramid Networks (PIIP). Specifically, PIIP uses pretrained models (ViTs or CNNs) as branches to process multi-scale images, where images of higher resolutions are processed by smaller network branches to balance computational cost and performance. To integrate information from different spatial scales, we further propose a novel cross-branch feature interaction mechanism. To validate PIIP, we apply it to various perception models and a representative multimodal large language model called LLaVA, and conduct extensive experiments on various tasks such as object detection, segmentation, image classification and multimodal understanding. PIIP achieves superior performance compared to single-branch and existing multi-resolution approaches with lower computational cost. When applied to InternViT-6B, a large-scale vision foundation model, PIIP can improve its performance by 1%-2% on detection and segmentation with only 40%-60% of the original computation, finally achieving 60.0 box AP on MS COCO and 59.7 mIoU on ADE20K. For multimodal understanding, our PIIP-LLaVA achieves 73.0% accuracy on TextVQA and 74.5% on MMBench with only 2.8M training data. Our code is released at https://github.com/OpenGVLab/PIIP.
☆ BMIP: Bi-directional Modality Interaction Prompt Learning for VLM
Vision-language models (VLMs) have exhibited remarkable generalization capabilities, and prompt learning for VLMs has attracted great attention for the ability to adapt pre-trained VLMs to specific downstream tasks. However, existing studies mainly focus on single-modal prompts or uni-directional modality interaction, overlooking the powerful alignment effects resulting from the interaction between the vision and language modalities. To this end, we propose a novel prompt learning method called $\underline{\textbf{B}}i-directional \underline{\textbf{M}}odality \underline{\textbf{I}}nteraction \underline{\textbf{P}}rompt (BMIP)$, which dynamically weights bi-modal information through learning the information of the attention layer, enhancing trainability and inter-modal consistency compared to simple information aggregation methods. To evaluate the effectiveness of prompt learning methods, we propose a more realistic evaluation paradigm called open-world generalization complementing the widely adopted cross-dataset transfer and domain generalization tasks. Comprehensive experiments on various datasets reveal that BMIP not only outperforms current state-of-the-art methods across all three evaluation paradigms but is also flexible enough to be combined with other prompt-based methods for consistent performance enhancement.
☆ PSReg: Prior-guided Sparse Mixture of Experts for Point Cloud Registration AAAI 2025
The discriminative feature is crucial for point cloud registration. Recent methods improve the feature discriminative by distinguishing between non-overlapping and overlapping region points. However, they still face challenges in distinguishing the ambiguous structures in the overlapping regions. Therefore, the ambiguous features they extracted resulted in a significant number of outlier matches from overlapping regions. To solve this problem, we propose a prior-guided SMoE-based registration method to improve the feature distinctiveness by dispatching the potential correspondences to the same experts. Specifically, we propose a prior-guided SMoE module by fusing prior overlap and potential correspondence embeddings for routing, assigning tokens to the most suitable experts for processing. In addition, we propose a registration framework by a specific combination of Transformer layer and prior-guided SMoE module. The proposed method not only pays attention to the importance of locating the overlapping areas of point clouds, but also commits to finding more accurate correspondences in overlapping areas. Our extensive experiments demonstrate the effectiveness of our method, achieving state-of-the-art registration recall (95.7\%/79.3\%) on the 3DMatch/3DLoMatch benchmark. Moreover, we also test the performance on ModelNet40 and demonstrate excellent performance.
comment: Accepted by AAAI 2025
☆ Automotive Elevation Mapping with Interferometric Synthetic Aperture Radar
Radar is a low-cost and ubiquitous automotive sensor, but is limited by array resolution and sensitivity when performing direction of arrival analysis. Synthetic Aperture Radar (SAR) is a class of techniques to improve azimuth resolution and sensitivity for radar. Interferometric SAR (InSAR) can be used to extract elevation from the variations in phase measurements in SAR images. Utilizing InSAR we show that a typical, low-resolution radar array mounted on a vehicle can be used to accurately localize detections in 3D space for both urban and agricultural environments. We generate point clouds in each environment by combining InSAR with a signal processing scheme tailored to automotive driving. This low-compute approach allows radar to be used as a primary sensor to map fine details in complex driving environments, and be used to make autonomous perception decisions.
comment: 9 pages, 6 figures
☆ FLAVARS: A Multimodal Foundational Language and Vision Alignment Model for Remote Sensing
Remote sensing imagery is dense with objects and contextual visual information. There is a recent trend to combine paired satellite images and text captions for pretraining performant encoders for downstream tasks. However, while contrastive image-text methods like CLIP enable vision-language alignment and zero-shot classification ability, vision-only downstream performance tends to degrade compared to image-only pretraining, such as MAE. In this paper, we propose FLAVARS, a pretraining method that combines the best of both contrastive learning and masked modeling, along with geospatial alignment via contrastive location encoding. We find that FLAVARS significantly outperforms a baseline of SkyCLIP for vision-only tasks such as KNN classification and semantic segmentation, +6\% mIOU on SpaceNet1, while retaining the ability to perform zero-shot classification, unlike MAE pretrained methods.
☆ Benchmarking Classical, Deep, and Generative Models for Human Activity Recognition
Human Activity Recognition (HAR) has gained significant importance with the growing use of sensor-equipped devices and large datasets. This paper evaluates the performance of three categories of models : classical machine learning, deep learning architectures, and Restricted Boltzmann Machines (RBMs) using five key benchmark datasets of HAR (UCI-HAR, OPPORTUNITY, PAMAP2, WISDM, and Berkeley MHAD). We assess various models, including Decision Trees, Random Forests, Convolutional Neural Networks (CNN), and Deep Belief Networks (DBNs), using metrics such as accuracy, precision, recall, and F1-score for a comprehensive comparison. The results show that CNN models offer superior performance across all datasets, especially on the Berkeley MHAD. Classical models like Random Forest do well on smaller datasets but face challenges with larger, more complex data. RBM-based models also show notable potential, particularly for feature learning. This paper offers a detailed comparison to help researchers choose the most suitable model for HAR tasks.
comment: 48 pages, 21 Figures
☆ Detecting Contextual Anomalies by Discovering Consistent Spatial Regions
We describe a method for modeling spatial context to enable video anomaly detection. The main idea is to discover regions that share similar object-level activities by clustering joint object attributes using Gaussian mixture models. We demonstrate that this straightforward approach, using orders of magnitude fewer parameters than competing models, achieves state-of-the-art performance in the challenging spatial-context-dependent Street Scene dataset. As a side benefit, the high-resolution discovered regions learned by the model also provide explainable normalcy maps for human operators without the need for any pre-trained segmentation model.
☆ Predicting Performance of Object Detection Models in Electron Microscopy Using Random Forests
Quantifying prediction uncertainty when applying object detection models to new, unlabeled datasets is critical in applied machine learning. This study introduces an approach to estimate the performance of deep learning-based object detection models for quantifying defects in transmission electron microscopy (TEM) images, focusing on detecting irradiation-induced cavities in TEM images of metal alloys. We developed a random forest regression model that predicts the object detection F1 score, a statistical metric used to evaluate the ability to accurately locate and classify objects of interest. The random forest model uses features extracted from the predictions of the object detection model whose uncertainty is being quantified, enabling fast prediction on new, unlabeled images. The mean absolute error (MAE) for predicting F1 of the trained model on test data is 0.09, and the $R^2$ score is 0.77, indicating there is a significant correlation between the random forest regression model predicted and true defect detection F1 scores. The approach is shown to be robust across three distinct TEM image datasets with varying imaging and material domains. Our approach enables users to estimate the reliability of a defect detection and segmentation model predictions and assess the applicability of the model to their specific datasets, providing valuable information about possible domain shifts and whether the model needs to be fine-tuned or trained on additional data to be maximally effective for the desired use case.
comment: 14 pages, 9 figures, 3 tables
☆ Towards Zero-Shot & Explainable Video Description by Reasoning over Graphs of Events in Space and Time
In the current era of Machine Learning, Transformers have become the de facto approach across a variety of domains, such as computer vision and natural language processing. Transformer-based solutions are the backbone of current state-of-the-art methods for language generation, image and video classification, segmentation, action and object recognition, among many others. Interestingly enough, while these state-of-the-art methods produce impressive results in their respective domains, the problem of understanding the relationship between vision and language is still beyond our reach. In this work, we propose a common ground between vision and language based on events in space and time in an explainable and programmatic way, to connect learning-based vision and language state of the art models and provide a solution to the long standing problem of describing videos in natural language. We validate that our algorithmic approach is able to generate coherent, rich and relevant textual descriptions on videos collected from a variety of datasets, using both standard metrics (e.g. Bleu, ROUGE) and the modern LLM-as-a-Jury approach.
☆ RWKV-UNet: Improving UNet with Long-Range Cooperation for Effective Medical Image Segmentation
In recent years, there have been significant advancements in deep learning for medical image analysis, especially with convolutional neural networks (CNNs) and transformer models. However, CNNs face limitations in capturing long-range dependencies while transformers suffer high computational complexities. To address this, we propose RWKV-UNet, a novel model that integrates the RWKV (Receptance Weighted Key Value) structure into the U-Net architecture. This integration enhances the model's ability to capture long-range dependencies and improve contextual understanding, which is crucial for accurate medical image segmentation. We build a strong encoder with developed inverted residual RWKV (IR-RWKV) blocks combining CNNs and RWKVs. We also propose a Cross-Channel Mix (CCM) module to improve skip connections with multi-scale feature fusion, achieving global channel information integration. Experiments on benchmark datasets, including Synapse, ACDC, BUSI, CVC-ClinicDB, CVC-ColonDB, Kvasir-SEG, ISIC 2017 and GLAS show that RWKV-UNet achieves state-of-the-art performance on various types of medical image segmentation. Additionally, smaller variants, RWKV-UNet-S and RWKV-UNet-T, balance accuracy and computational efficiency, making them suitable for broader clinical applications.
☆ Vchitect-2.0: Parallel Transformer for Scaling Up Video Diffusion Models
We present Vchitect-2.0, a parallel transformer architecture designed to scale up video diffusion models for large-scale text-to-video generation. The overall Vchitect-2.0 system has several key designs. (1) By introducing a novel Multimodal Diffusion Block, our approach achieves consistent alignment between text descriptions and generated video frames, while maintaining temporal coherence across sequences. (2) To overcome memory and computational bottlenecks, we propose a Memory-efficient Training framework that incorporates hybrid parallelism and other memory reduction techniques, enabling efficient training of long video sequences on distributed systems. (3) Additionally, our enhanced data processing pipeline ensures the creation of Vchitect T2V DataVerse, a high-quality million-scale training dataset through rigorous annotation and aesthetic evaluation. Extensive benchmarking demonstrates that Vchitect-2.0 outperforms existing methods in video quality, training efficiency, and scalability, serving as a suitable base for high-fidelity video generation.
☆ Poseidon: A ViT-based Architecture for Multi-Frame Pose Estimation with Adaptive Frame Weighting and Multi-Scale Feature Fusion
Human pose estimation, a vital task in computer vision, involves detecting and localising human joints in images and videos. While single-frame pose estimation has seen significant progress, it often fails to capture the temporal dynamics for understanding complex, continuous movements. We propose Poseidon, a novel multi-frame pose estimation architecture that extends the ViTPose model by integrating temporal information for enhanced accuracy and robustness to address these limitations. Poseidon introduces key innovations: (1) an Adaptive Frame Weighting (AFW) mechanism that dynamically prioritises frames based on their relevance, ensuring that the model focuses on the most informative data; (2) a Multi-Scale Feature Fusion (MSFF) module that aggregates features from different backbone layers to capture both fine-grained details and high-level semantics; and (3) a Cross-Attention module for effective information exchange between central and contextual frames, enhancing the model's temporal coherence. The proposed architecture improves performance in complex video scenarios and offers scalability and computational efficiency suitable for real-world applications. Our approach achieves state-of-the-art performance on the PoseTrack21 and PoseTrack18 datasets, achieving mAP scores of 88.3 and 87.8, respectively, outperforming existing methods.
☆ FARE: A Deep Learning-Based Framework for Radar-based Face Recognition and Out-of-distribution Detection ICASSP 2025
In this work, we propose a novel pipeline for face recognition and out-of-distribution (OOD) detection using short-range FMCW radar. The proposed system utilizes Range-Doppler and micro Range-Doppler Images. The architecture features a primary path (PP) responsible for the classification of in-distribution (ID) faces, complemented by intermediate paths (IPs) dedicated to OOD detection. The network is trained in two stages: first, the PP is trained using triplet loss to optimize ID face classification. In the second stage, the PP is frozen, and the IPs-comprising simple linear autoencoder networks-are trained specifically for OOD detection. Using our dataset generated with a 60 GHz FMCW radar, our method achieves an ID classification accuracy of 99.30% and an OOD detection AUROC of 96.91%.
comment: Accepted at ICASSP 2025
☆ Cross-Modal Transferable Image-to-Video Attack on Video Quality Metrics
Recent studies have revealed that modern image and video quality assessment (IQA/VQA) metrics are vulnerable to adversarial attacks. An attacker can manipulate a video through preprocessing to artificially increase its quality score according to a certain metric, despite no actual improvement in visual quality. Most of the attacks studied in the literature are white-box attacks, while black-box attacks in the context of VQA have received less attention. Moreover, some research indicates a lack of transferability of adversarial examples generated for one model to another when applied to VQA. In this paper, we propose a cross-modal attack method, IC2VQA, aimed at exploring the vulnerabilities of modern VQA models. This approach is motivated by the observation that the low-level feature spaces of images and videos are similar. We investigate the transferability of adversarial perturbations across different modalities; specifically, we analyze how adversarial perturbations generated on a white-box IQA model with an additional CLIP module can effectively target a VQA model. The addition of the CLIP module serves as a valuable aid in increasing transferability, as the CLIP model is known for its effective capture of low-level semantics. Extensive experiments demonstrate that IC2VQA achieves a high success rate in attacking three black-box VQA models. We compare our method with existing black-box attack strategies, highlighting its superiority in terms of attack success within the same number of iterations and levels of attack strength. We believe that the proposed method will contribute to the deeper analysis of robust VQA metrics.
comment: Accepted for VISAPP 2025
☆ BiDepth Multimodal Neural Network: Bidirectional Depth Deep Learning Arcitecture for Spatial-Temporal Prediction
Accurate prediction of spatial-temporal (ST) information in dynamic systems, such as urban mobility and weather patterns, is a crucial yet challenging problem. The complexity stems from the intricate interplay between spatial proximity and temporal relevance, where both long-term trends and short-term fluctuations are present in convoluted patterns. Existing approaches, including traditional statistical methods and conventional neural networks, may provide inaccurate results due to the lack of an effective mechanism that simultaneously incorporates information at variable temporal depths while maintaining spatial context, resulting in a trade-off between comprehensive long-term historical analysis and responsiveness to short-term new information. To bridge this gap, this paper proposes the BiDepth Multimodal Neural Network (BDMNN) with bidirectional depth modulation that enables a comprehensive understanding of both long-term seasonality and short-term fluctuations, adapting to the complex ST context. Case studies with real-world public data demonstrate significant improvements in prediction accuracy, with a 12% reduction in Mean Squared Error for urban traffic prediction and a 15% improvement in rain precipitation forecasting compared to state-of-the-art benchmarks, without demanding extra computational resources.
comment: This paper has been submitted to Applied Intelligence for review
☆ Leveraging 2D Masked Reconstruction for Domain Adaptation of 3D Pose Estimation
RGB-based 3D pose estimation methods have been successful with the development of deep learning and the emergence of high-quality 3D pose datasets. However, most existing methods do not operate well for testing images whose distribution is far from that of training data. However, most existing methods do not operate well for testing images whose distribution is far from that of training data. This problem might be alleviated by involving diverse data during training, however it is non-trivial to collect such diverse data with corresponding labels (i.e. 3D pose). In this paper, we introduced an unsupervised domain adaptation framework for 3D pose estimation that utilizes the unlabeled data in addition to labeled data via masked image modeling (MIM) framework. Foreground-centric reconstruction and attention regularization are further proposed to increase the effectiveness of unlabeled data usage. Experiments are conducted on the various datasets in human and hand pose estimation tasks, especially using the cross-domain scenario. We demonstrated the effectiveness of ours by achieving the state-of-the-art accuracy on all datasets.
comment: 16 pages, 7 figures
☆ 3D Gaussian Splatting with Normal Information for Mesh Extraction and Improved Rendering ICASSP 2025
Differentiable 3D Gaussian splatting has emerged as an efficient and flexible rendering technique for representing complex scenes from a collection of 2D views and enabling high-quality real-time novel-view synthesis. However, its reliance on photometric losses can lead to imprecisely reconstructed geometry and extracted meshes, especially in regions with high curvature or fine detail. We propose a novel regularization method using the gradients of a signed distance function estimated from the Gaussians, to improve the quality of rendering while also extracting a surface mesh. The regularizing normal supervision facilitates better rendering and mesh reconstruction, which is crucial for downstream applications in video generation, animation, AR-VR and gaming. We demonstrate the effectiveness of our approach on datasets such as Mip-NeRF360, Tanks and Temples, and Deep-Blending. Our method scores higher on photorealism metrics compared to other mesh extracting rendering methods without compromising mesh quality.
comment: ICASSP 2025: Workshop on Generative Data Augmentation for Real-World Signal Processing Applications
Rate-In: Information-Driven Adaptive Dropout Rates for Improved Inference-Time Uncertainty Estimation
Accurate uncertainty estimation is crucial for deploying neural networks in risk-sensitive applications such as medical diagnosis. Monte Carlo Dropout is a widely used technique for approximating predictive uncertainty by performing stochastic forward passes with dropout during inference. However, using static dropout rates across all layers and inputs can lead to suboptimal uncertainty estimates, as it fails to adapt to the varying characteristics of individual inputs and network layers. Existing approaches optimize dropout rates during training using labeled data, resulting in fixed inference-time parameters that cannot adjust to new data distributions, compromising uncertainty estimates in Monte Carlo simulations. In this paper, we propose Rate-In, an algorithm that dynamically adjusts dropout rates during inference by quantifying the information loss induced by dropout in each layer's feature maps. By treating dropout as controlled noise injection and leveraging information-theoretic principles, Rate-In adapts dropout rates per layer and per input instance without requiring ground truth labels. By quantifying the functional information loss in feature maps, we adaptively tune dropout rates to maintain perceptual quality across diverse medical imaging tasks and architectural configurations. Our extensive empirical study on synthetic data and real-world medical imaging tasks demonstrates that Rate-In improves calibration and sharpens uncertainty estimates compared to fixed or heuristic dropout rates without compromising predictive performance. Rate-In offers a practical, unsupervised, inference-time approach to optimizing dropout for more reliable predictive uncertainty estimation in critical applications.
comment: Updated author affiliation
♻ ☆ Gaussian Eigen Models for Human Heads
Current personalized neural head avatars face a trade-off: lightweight models lack detail and realism, while high-quality, animatable avatars require significant computational resources, making them unsuitable for commodity devices. To address this gap, we introduce Gaussian Eigen Models (GEM), which provide high-quality, lightweight, and easily controllable head avatars. GEM utilizes 3D Gaussian primitives for representing the appearance combined with Gaussian splatting for rendering. Building on the success of mesh-based 3D morphable face models (3DMM), we define GEM as an ensemble of linear eigenbases for representing the head appearance of a specific subject. In particular, we construct linear bases to represent the position, scale, rotation, and opacity of the 3D Gaussians. This allows us to efficiently generate Gaussian primitives of a specific head shape by a linear combination of the basis vectors, only requiring a low-dimensional parameter vector that contains the respective coefficients. We propose to construct these linear bases (GEM) by distilling high-quality compute-intense CNN-based Gaussian avatar models that can generate expression-dependent appearance changes like wrinkles. These high-quality models are trained on multi-view videos of a subject and are distilled using a series of principal component analyses. Once we have obtained the bases that represent the animatable appearance space of a specific human, we learn a regressor that takes a single RGB image as input and predicts the low-dimensional parameter vector that corresponds to the shown facial expression. In a series of experiments, we compare GEM's self-reenactment and cross-person reenactment results to state-of-the-art 3D avatar methods, demonstrating GEM's higher visual quality and better generalization to new expressions.
comment: https://zielon.github.io/gem/
♻ ☆ A Multi-Modal Approach for Face Anti-Spoofing in Non-Calibrated Systems using Disparity Maps
Face recognition technologies are increasingly used in various applications, yet they are vulnerable to face spoofing attacks. These spoofing attacks often involve unique 3D structures, such as printed papers or mobile device screens. Although stereo-depth cameras can detect such attacks effectively, their high-cost limits their widespread adoption. Conversely, two-sensor systems without extrinsic calibration offer a cost-effective alternative but are unable to calculate depth using stereo techniques. In this work, we propose a method to overcome this challenge by leveraging facial attributes to derive disparity information and estimate relative depth for anti-spoofing purposes, using non-calibrated systems. We introduce a multi-modal anti-spoofing model, coined Disparity Model, that incorporates created disparity maps as a third modality alongside the two original sensor modalities. We demonstrate the effectiveness of the Disparity Model in countering various spoof attacks using a comprehensive dataset collected from the Intel RealSense ID Solution F455. Our method outperformed existing methods in the literature, achieving an Equal Error Rate (EER) of 1.71% and a False Negative Rate (FNR) of 2.77% at a False Positive Rate (FPR) of 1%. These errors are lower by 2.45% and 7.94% than the errors of the best comparison method, respectively. Additionally, we introduce a model ensemble that addresses 3D spoof attacks as well, achieving an EER of 2.04% and an FNR of 3.83% at an FPR of 1%. Overall, our work provides a state-of-the-art solution for the challenging task of anti-spoofing in non-calibrated systems that lack depth information.
♻ ☆ RMem: Restricted Memory Banks Improve Video Object Segmentation CVPR 2024
With recent video object segmentation (VOS) benchmarks evolving to challenging scenarios, we revisit a simple but overlooked strategy: restricting the size of memory banks. This diverges from the prevalent practice of expanding memory banks to accommodate extensive historical information. Our specially designed "memory deciphering" study offers a pivotal insight underpinning such a strategy: expanding memory banks, while seemingly beneficial, actually increases the difficulty for VOS modules to decode relevant features due to the confusion from redundant information. By restricting memory banks to a limited number of essential frames, we achieve a notable improvement in VOS accuracy. This process balances the importance and freshness of frames to maintain an informative memory bank within a bounded capacity. Additionally, restricted memory banks reduce the training-inference discrepancy in memory lengths compared with continuous expansion. This fosters new opportunities in temporal reasoning and enables us to introduce the previously overlooked "temporal positional embedding." Finally, our insights are embodied in "RMem" ("R" for restricted), a simple yet effective VOS modification that excels at challenging VOS scenarios and establishes new state of the art for object state changes (on the VOST dataset) and long videos (on the Long Videos dataset). Our code and demo are available at https://restricted-memory.github.io/.
comment: CVPR 2024, Project Page: https://restricted-memory.github.io/
♻ ☆ FaVoR: Features via Voxel Rendering for Camera Relocalization WACV
Camera relocalization methods range from dense image alignment to direct camera pose regression from a query image. Among these, sparse feature matching stands out as an efficient, versatile, and generally lightweight approach with numerous applications. However, feature-based methods often struggle with significant viewpoint and appearance changes, leading to matching failures and inaccurate pose estimates. To overcome this limitation, we propose a novel approach that leverages a globally sparse yet locally dense 3D representation of 2D features. By tracking and triangulating landmarks over a sequence of frames, we construct a sparse voxel map optimized to render image patch descriptors observed during tracking. Given an initial pose estimate, we first synthesize descriptors from the voxels using volumetric rendering and then perform feature matching to estimate the camera pose. This methodology enables the generation of descriptors for unseen views, enhancing robustness to view changes. We extensively evaluate our method on the 7-Scenes and Cambridge Landmarks datasets. Our results show that our method significantly outperforms existing state-of-the-art feature representation techniques in indoor environments, achieving up to a 39% improvement in median translation error. Additionally, our approach yields comparable results to other methods for outdoor scenarios while maintaining lower memory and computational costs.
comment: Accepted to the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Tucson, Arizona, US, Feb 28-Mar 4, 2025
♻ ☆ Vid2Sim: Realistic and Interactive Simulation from Video for Urban Navigation
Sim-to-real gap has long posed a significant challenge for robot learning in simulation, preventing the deployment of learned models in the real world. Previous work has primarily focused on domain randomization and system identification to mitigate this gap. However, these methods are often limited by the inherent constraints of the simulation and graphics engines. In this work, we propose Vid2Sim, a novel framework that effectively bridges the sim2real gap through a scalable and cost-efficient real2sim pipeline for neural 3D scene reconstruction and simulation. Given a monocular video as input, Vid2Sim can generate photorealistic and physically interactable 3D simulation environments to enable the reinforcement learning of visual navigation agents in complex urban environments. Extensive experiments demonstrate that Vid2Sim significantly improves the performance of urban navigation in the digital twins and real world by 31.2% and 68.3% in success rate compared with agents trained with prior simulation methods.
comment: Project page: https://metadriverse.github.io/vid2sim/
♻ ☆ Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models
We present Deep Compression Autoencoder (DC-AE), a new family of autoencoder models for accelerating high-resolution diffusion models. Existing autoencoder models have demonstrated impressive results at a moderate spatial compression ratio (e.g., 8x), but fail to maintain satisfactory reconstruction accuracy for high spatial compression ratios (e.g., 64x). We address this challenge by introducing two key techniques: (1) Residual Autoencoding, where we design our models to learn residuals based on the space-to-channel transformed features to alleviate the optimization difficulty of high spatial-compression autoencoders; (2) Decoupled High-Resolution Adaptation, an efficient decoupled three-phases training strategy for mitigating the generalization penalty of high spatial-compression autoencoders. With these designs, we improve the autoencoder's spatial compression ratio up to 128 while maintaining the reconstruction quality. Applying our DC-AE to latent diffusion models, we achieve significant speedup without accuracy drop. For example, on ImageNet 512x512, our DC-AE provides 19.1x inference speedup and 17.9x training speedup on H100 GPU for UViT-H while achieving a better FID, compared with the widely used SD-VAE-f8 autoencoder. Our code is available at https://github.com/mit-han-lab/efficientvit.
comment: Preprint. First two authors contributed equally to this work. Update: add USiT (UViT+SiT sampler) results
♻ ☆ Scaling White-Box Transformers for Vision
CRATE, a white-box transformer architecture designed to learn compressed and sparse representations, offers an intriguing alternative to standard vision transformers (ViTs) due to its inherent mathematical interpretability. Despite extensive investigations into the scaling behaviors of language and vision transformers, the scalability of CRATE remains an open question which this paper aims to address. Specifically, we propose CRATE-$\alpha$, featuring strategic yet minimal modifications to the sparse coding block in the CRATE architecture design, and a light training recipe designed to improve the scalability of CRATE. Through extensive experiments, we demonstrate that CRATE-$\alpha$ can effectively scale with larger model sizes and datasets. For example, our CRATE-$\alpha$-B substantially outperforms the prior best CRATE-B model accuracy on ImageNet classification by 3.7%, achieving an accuracy of 83.2%. Meanwhile, when scaling further, our CRATE-$\alpha$-L obtains an ImageNet classification accuracy of 85.1%. More notably, these model performance improvements are achieved while preserving, and potentially even enhancing the interpretability of learned CRATE models, as we demonstrate through showing that the learned token representations of increasingly larger trained CRATE-$\alpha$ models yield increasingly higher-quality unsupervised object segmentation of images. The project page is https://rayjryang.github.io/CRATE-alpha/.
comment: project page: https://rayjryang.github.io/CRATE-alpha/
♻ ☆ A Comprehensive Survey of Foundation Models in Medicine
Foundation models (FMs) are large-scale deep learning models that are developed using large datasets and self-supervised learning methods. These models serve as a base for different downstream tasks, including healthcare. FMs have been adopted with great success across various domains within healthcare. Existing healthcare-based surveys have not yet included all of these domains. Therefore, we provide a detailed survey of FMs in healthcare. We focus on the history, learning strategies, flagship models, applications, and challenges of FMs. We explore how FMs such as the BERT and GPT families are reshaping various healthcare domains, including clinical large language models, medical image analysis, and omics. Furthermore, we provide a detailed taxonomy of healthcare applications facilitated by FMs, such as clinical NLP, medical computer vision, graph learning, and other biology-related tasks. Despite the promising opportunities FMs provide, they also have several associated challenges, which are explained in detail. We also outline open research issues and potential lessons learned to provide researchers and practitioners with insights into the capabilities of FMs in healthcare to advance their deployment and mitigate associated risks.
comment: Currently under review in IEEE REVIEWS IN BIOMEDICAL ENGINEERING
♻ ☆ Text-guided Image Restoration and Semantic Enhancement for Text-to-Image Person Retrieval
The goal of Text-to-Image Person Retrieval (TIPR) is to retrieve specific person images according to the given textual descriptions. A primary challenge in this task is bridging the substantial representational gap between visual and textual modalities. The prevailing methods map texts and images into unified embedding space for matching, while the intricate semantic correspondences between texts and images are still not effectively constructed. To address this issue, we propose a novel TIPR framework to build fine-grained interactions and alignment between person images and the corresponding texts. Specifically, via fine-tuning the Contrastive Language-Image Pre-training (CLIP) model, a visual-textual dual encoder is firstly constructed, to preliminarily align the image and text features. Secondly, a Text-guided Image Restoration (TIR) auxiliary task is proposed to map abstract textual entities to specific image regions, improving the alignment between local textual and visual embeddings. Additionally, a cross-modal triplet loss is presented to handle hard samples, and further enhance the model's discriminability for minor differences. Moreover, a pruning-based text data augmentation approach is proposed to enhance focus on essential elements in descriptions, thereby avoiding excessive model attention to less significant information. The experimental results show our proposed method outperforms state-of-the-art methods on three popular benchmark datasets, and the code will be made publicly available at https://github.com/Delong-liu-bupt/SEN.
comment: The paper was withdrawn due to a dispute among the authors regarding the content of the article
♻ ☆ Relaxed Rotational Equivariance via $G$-Biases in Vision
Group Equivariant Convolution (GConv) can capture rotational equivariance from original data. It assumes uniform and strict rotational equivariance across all features as the transformations under the specific group. However, the presentation or distribution of real-world data rarely conforms to strict rotational equivariance, commonly referred to as Rotational Symmetry-Breaking (RSB) in the system or dataset, making GConv unable to adapt effectively to this phenomenon. Motivated by this, we propose a simple but highly effective method to address this problem, which utilizes a set of learnable biases called $G$-Biases under the group order to break strict group constraints and then achieve a Relaxed Rotational Equivariant Convolution (RREConv). To validate the efficiency of RREConv, we conduct extensive ablation experiments on the discrete rotational group $\mathcal{C}_n$. Experiments demonstrate that the proposed RREConv-based methods achieve excellent performance compared to existing GConv-based methods in both classification and 2D object detection tasks on the natural image datasets.
♻ ☆ Feedback-driven object detection and iterative model improvement
Automated object detection has become increasingly valuable across diverse applications, yet efficient, high-quality annotation remains a persistent challenge. In this paper, we present the development and evaluation of a platform designed to interactively improve object detection models. The platform allows uploading and annotating images as well as fine-tuning object detection models. Users can then manually review and refine annotations, further creating improved snapshots that are used for automatic object detection on subsequent image uploads - a process we refer to as semi-automatic annotation resulting in a significant gain in annotation efficiency. Whereas iterative refinement of model results to speed up annotation has become common practice, we are the first to quantitatively evaluate its benefits with respect to time, effort, and interaction savings. Our experimental results show clear evidence for a significant time reduction of up to 53% for semi-automatic compared to manual annotation. Importantly, these efficiency gains did not compromise annotation quality, while matching or occasionally even exceeding the accuracy of manual annotations. These findings demonstrate the potential of our lightweight annotation platform for creating high-quality object detection datasets and provide best practices to guide future development of annotation platforms. The platform is open-source, with the frontend and backend repositories available on GitHub (https://github.com/ml-lab-htw/iterative-annotate). To support the understanding of our labeling process, we have created an explanatory video demonstrating the methodology using microscopy images of E. coli bacteria as an example. The video is available on YouTube (https://www.youtube.com/watch?v=CM9uhE8NN5E).
comment: AI4EA24
♻ ☆ ORFormer: Occlusion-Robust Transformer for Accurate Facial Landmark Detection WACV 2025
Although facial landmark detection (FLD) has gained significant progress, existing FLD methods still suffer from performance drops on partially non-visible faces, such as faces with occlusions or under extreme lighting conditions or poses. To address this issue, we introduce ORFormer, a novel transformer-based method that can detect non-visible regions and recover their missing features from visible parts. Specifically, ORFormer associates each image patch token with one additional learnable token called the messenger token. The messenger token aggregates features from all but its patch. This way, the consensus between a patch and other patches can be assessed by referring to the similarity between its regular and messenger embeddings, enabling non-visible region identification. Our method then recovers occluded patches with features aggregated by the messenger tokens. Leveraging the recovered features, ORFormer compiles high-quality heatmaps for the downstream FLD task. Extensive experiments show that our method generates heatmaps resilient to partial occlusions. By integrating the resultant heatmaps into existing FLD methods, our method performs favorably against the state of the arts on challenging datasets such as WFLW and COFW.
comment: WACV 2025 Project Link: https://ben0919.github.io/ORFormer/
♻ ☆ Diversified Augmentation with Domain Adaptation for Debiased Video Temporal Grounding ICASSP 2025
Temporal sentence grounding in videos (TSGV) faces challenges due to public TSGV datasets containing significant temporal biases, which are attributed to the uneven temporal distributions of target moments. Existing methods generate augmented videos, where target moments are forced to have varying temporal locations. However, since the video lengths of the given datasets have small variations, only changing the temporal locations results in poor generalization ability in videos with varying lengths. In this paper, we propose a novel training framework complemented by diversified data augmentation and a domain discriminator. The data augmentation generates videos with various lengths and target moment locations to diversify temporal distributions. However, augmented videos inevitably exhibit distinct feature distributions which may introduce noise. To address this, we design a domain adaptation auxiliary task to diminish feature discrepancies between original and augmented videos. We also encourage the model to produce distinct predictions for videos with the same text queries but different moment locations to promote debiased training. Experiments on Charades-CD and ActivityNet-CD datasets demonstrate the effectiveness and generalization abilities of our method in multiple grounding structures, achieving state-of-the-art results.
comment: Accepted by ICASSP 2025
♻ ☆ MSCViT: A Small-size ViT architecture with Multi-Scale Self-Attention Mechanism for Tiny Datasets
Vision Transformer (ViT) has demonstrated significant potential in various vision tasks due to its strong ability in modelling long-range dependencies. However, such success is largely fueled by training on massive samples. In real applications, the large-scale datasets are not always available, and ViT performs worse than Convolutional Neural Networks (CNNs) if it is only trained on small scale dataset (called tiny dataset), since it requires large amount of training data to ensure its representational capacity. In this paper, a small-size ViT architecture with multi-scale self-attention mechanism and convolution blocks is presented (dubbed MSCViT) to model different scales of attention at each layer. Firstly, we introduced wavelet convolution, which selectively combines the high-frequency components obtained by frequency division with our convolution channel to extract local features. Then, a lightweight multi-head attention module is developed to reduce the number of tokens and computational costs. Finally, the positional encoding (PE) in the backbone is replaced by a local feature extraction module. Compared with the original ViT, it is parameter-efficient and is particularly suitable for tiny datasets. Extensive experiments have been conducted on tiny datasets, in which our model achieves an accuracy of 84.68% on CIFAR-100 with 14.0M parameters and 2.5 GFLOPs, without pre-training on large datasets.
♻ ☆ WINE: Wavelet-Guided GAN Inversion and Editing for High-Fidelity Refinement
Recent advanced GAN inversion models aim to convey high-fidelity information from original images to generators through methods using generator tuning or high-dimensional feature learning. Despite these efforts, accurately reconstructing image-specific details remains as a challenge due to the inherent limitations both in terms of training and structural aspects, leading to a bias towards low-frequency information. In this paper, we look into the widely used pixel loss in GAN inversion, revealing its predominant focus on the reconstruction of low-frequency features. We then propose WINE, a Wavelet-guided GAN Inversion aNd Editing model, which transfers the high-frequency information through wavelet coefficients via newly proposed wavelet loss and wavelet fusion scheme. Notably, WINE is the first attempt to interpret GAN inversion in the frequency domain. Our experimental results showcase the precision of WINE in preserving high-frequency details and enhancing image quality. Even in editing scenarios, WINE outperforms existing state-of-the-art GAN inversion models with a fine balance between editability and reconstruction quality.
♻ ☆ Generalized and Efficient 2D Gaussian Splatting for Arbitrary-scale Super-Resolution
Equipped with the continuous representation capability of Multi-Layer Perceptron (MLP), Implicit Neural Representation (INR) has been successfully employed for Arbitrary-scale Super-Resolution (ASR). However, the limited receptive field of the linear layers in MLP restricts the representation capability of INR, while it is computationally expensive to query the MLP numerous times to render each pixel. Recently, Gaussian Splatting (GS) has shown its advantages over INR in both visual quality and rendering speed in 3D tasks, which motivates us to explore whether GS can be employed for the ASR task. However, directly applying GS to ASR is exceptionally challenging because the original GS is an optimization-based method through overfitting each single scene, while in ASR we aim to learn a single model that can generalize to different images and scaling factors. We overcome these challenges by developing two novel techniques. Firstly, to generalize GS for ASR, we elaborately design an architecture to predict the corresponding image-conditioned Gaussians of the input low-resolution image in a feed-forward manner. Secondly, we implement an efficient differentiable 2D GPU/CUDA-based scale-aware rasterization to render super-resolved images by sampling discrete RGB values from the predicted contiguous Gaussians. Via end-to-end training, our optimized network, namely GSASR, can perform ASR for any image and unseen scaling factors. Extensive experiments validate the effectiveness of our proposed method. The project page can be found at \url{https://mt-cly.github.io/GSASR.github.io/}.
♻ ☆ Dynamic Sub-graph Distillation for Robust Semi-supervised Continual Learning
Continual learning (CL) has shown promising results and comparable performance to learning at once in a fully supervised manner. However, CL strategies typically require a large number of labeled samples, making their real-life deployment challenging. In this work, we focus on semi-supervised continual learning (SSCL), where the model progressively learns from partially labeled data with unknown categories. We provide a comprehensive analysis of SSCL and demonstrate that unreliable distributions of unlabeled data lead to unstable training and refinement of the progressing stages. This problem severely impacts the performance of SSCL. To address the limitations, we propose a novel approach called Dynamic Sub-Graph Distillation (DSGD) for semi-supervised continual learning, which leverages both semantic and structural information to achieve more stable knowledge distillation on unlabeled data and exhibit robustness against distribution bias. Firstly, we formalize a general model of structural distillation and design a dynamic graph construction for the continual learning progress. Next, we define a structure distillation vector and design a dynamic sub-graph distillation algorithm, which enables end-to-end training and adaptability to scale up tasks. The entire proposed method is adaptable to various CL methods and supervision settings. Finally, experiments conducted on three datasets CIFAR10, CIFAR100, and ImageNet-100, with varying supervision ratios, demonstrate the effectiveness of our proposed approach in mitigating the catastrophic forgetting problem in semi-supervised continual learning scenarios.
♻ ☆ Less is More: The Influence of Pruning on the Explainability of CNNs
Over the last century, deep learning models have become the state-of-the-art for solving complex computer vision problems. These modern computer vision models have millions of parameters, which presents two major challenges: (1) the increased computational requirements hamper the deployment in resource-constrained environments, such as mobile or IoT devices, and (2) explaining the complex decisions of such networks to humans is challenging. Network pruning is a technical approach to reduce the complexity of models, where less important parameters are removed. The work presented in this paper investigates whether this reduction in technical complexity also helps with perceived explainability. To do so, we conducted a pre-study and two human-grounded experiments, assessing the effects of different pruning ratios on explainability. Overall, we evaluate four different compression rates (i.e., 2, 4, 8, and 32) with 37 500 tasks on Mechanical Turk. Results indicate that lower compression rates have a positive influence on explainability, while higher compression rates show negative effects. Furthermore, we were able to identify sweet spots that increase both the perceived explainability and the model's performance.
♻ ☆ Spurious Feature Eraser: Stabilizing Test-Time Adaptation for Vision-Language Foundation Model
Vision-language foundation models have exhibited remarkable success across a multitude of downstream tasks due to their scalability on extensive image-text paired data. However, these models also display significant limitations when applied to downstream tasks, such as fine-grained image classification, as a result of ``decision shortcuts'' that hinder their generalization capabilities. In this work, we find that the CLIP model possesses a rich set of features, encompassing both \textit{desired invariant causal features} and \textit{undesired decision shortcuts}. Moreover, the underperformance of CLIP on downstream tasks originates from its inability to effectively utilize pre-trained features in accordance with specific task requirements. To address this challenge, we propose a simple yet effective method, Spurious Feature Eraser (SEraser), to alleviate the decision shortcuts by erasing the spurious features. Specifically, we introduce a test-time prompt tuning paradigm that optimizes a learnable prompt, thereby compelling the model to exploit invariant features while disregarding decision shortcuts during the inference phase. The proposed method effectively alleviates excessive dependence on potentially misleading spurious information. We conduct comparative analysis of the proposed method against various approaches which validates the significant superiority.
♻ ☆ ImagiNet: A Multi-Content Benchmark for Synthetic Image Detection AAAI 2025
Recent generative models produce images with a level of authenticity that makes them nearly indistinguishable from real photos and artwork. Potential harmful use cases of these models, necessitate the creation of robust synthetic image detectors. However, current datasets in the field contain generated images with questionable quality or have examples from one predominant content type which leads to poor generalizability of the underlying detectors. We find that the curation of a balanced amount of high-resolution generated images across various content types is crucial for the generalizability of detectors, and introduce ImagiNet, a dataset of 200K examples, spanning four categories: photos, paintings, faces, and miscellaneous. Synthetic images in ImagiNet are produced with both open-source and proprietary generators, whereas real counterparts for each content type are collected from public datasets. The structure of ImagiNet allows for a two-track evaluation system: i) classification as real or synthetic and ii) identification of the generative model. To establish a strong baseline, we train a ResNet-50 model using a self-supervised contrastive objective (SelfCon) for each track which achieves evaluation AUC of up to 0.99 and balanced accuracy ranging from 86% to 95%, even under conditions that involve compression and resizing. The provided model is generalizable enough to achieve zero-shot state-of-the-art performance on previous synthetic detection benchmarks. We provide ablations to demonstrate the importance of content types and publish code and data.
comment: Workshop on Datasets and Evaluators of AI Safety, AAAI 2025
♻ ☆ Digi2Real: Bridging the Realism Gap in Synthetic Data Face Recognition via Foundation Models WACV 2025
The accuracy of face recognition systems has improved significantly in the past few years, thanks to the large amount of data collected and advancements in neural network architectures. However, these large-scale datasets are often collected without explicit consent, raising ethical and privacy concerns. To address this, there have been proposals to use synthetic datasets for training face recognition models. Yet, such models still rely on real data to train the generative models and generally exhibit inferior performance compared to those trained on real datasets. One of these datasets, DigiFace, uses a graphics pipeline to generate different identities and intra-class variations without using real data in model training. However, the performance of this approach is poor on face recognition benchmarks, possibly due to the lack of realism in the images generated by the graphics pipeline. In this work, we introduce a novel framework for realism transfer aimed at enhancing the realism of synthetically generated face images. Our method leverages the large-scale face foundation model, and we adapt the pipeline for realism enhancement. By integrating the controllable aspects of the graphics pipeline with our realism enhancement technique, we generate a large amount of realistic variations, combining the advantages of both approaches. Our empirical evaluations demonstrate that models trained using our enhanced dataset significantly improve the performance of face recognition systems over the baseline. The source code and dataset will be publicly accessible at the following link: https://www.idiap.ch/paper/digi2real
comment: The dataset would be available here: https://www.idiap.ch/paper/digi2real Accepted for Publication in WACV 2025
♻ ☆ MambaTalk: Efficient Holistic Gesture Synthesis with Selective State Space Models
Gesture synthesis is a vital realm of human-computer interaction, with wide-ranging applications across various fields like film, robotics, and virtual reality. Recent advancements have utilized the diffusion model and attention mechanisms to improve gesture synthesis. However, due to the high computational complexity of these techniques, generating long and diverse sequences with low latency remains a challenge. We explore the potential of state space models (SSMs) to address the challenge, implementing a two-stage modeling strategy with discrete motion priors to enhance the quality of gestures. Leveraging the foundational Mamba block, we introduce MambaTalk, enhancing gesture diversity and rhythm through multimodal integration. Extensive experiments demonstrate that our method matches or exceeds the performance of state-of-the-art models.
comment: NeurlPS 2024, Camera Ready
♻ ☆ Audio-Agent: Leveraging LLMs For Audio Generation, Editing and Composition
We introduce Audio-Agent, a multimodal framework for audio generation, editing and composition based on text or video inputs. Conventional approaches for text-to-audio (TTA) tasks often make single-pass inferences from text descriptions. While straightforward, this design struggles to produce high-quality audio when given complex text conditions. In our method, we utilize a pre-trained TTA diffusion network as the audio generation agent to work in tandem with GPT-4, which decomposes the text condition into atomic, specific instructions and calls the agent for audio generation. In doing so, Audio-Agent can generate high-quality audio that is closely aligned with the provided text or video exhibiting complex and multiple events, while supporting variable-length and variable-volume generation. For video-to-audio (VTA) tasks, most existing methods require training a timestamp detector to synchronize video events with the generated audio, a process that can be tedious and time-consuming. Instead, we propose a simpler approach by fine-tuning a pre-trained Large Language Model (LLM), e.g., Gemma2-2B-it, to obtain both semantic and temporal conditions that bridge the video and audio modality. Consequently, our framework contributes a comprehensive solution for both TTA and VTA tasks without substantial computational overhead in training.
♻ ☆ EventHallusion: Diagnosing Event Hallucinations in Video LLMs
Recently, Multimodal Large Language Models (MLLMs) have made significant progress in the video comprehension field. Despite remarkable content reasoning and instruction following capabilities they demonstrated, the hallucination problem of these VideoLLMs is less explored compared with its counterpart in the image domain. To mitigate this gap, we propose EventHallusion, a novel benchmark that focuses on assessing the VideoLLMs' hallucination toward event, the crux of video analysis. From a hallucination attribution perspective, our EventHallusion benchmark is curated to assess a VideoLLM's susceptibility toward language priors and vision-language biases. On the other hand, we also propose a simple yet effective method, called Temporal Contrastive Decoding (TCD), to tackle the hallucination problems of VideoLLMs. The proposed TCD method rectifies the model's bias toward its priors during the decoding stage by comparing the original video with a modified version, in which temporal cues are disrupted. Through comprehensive evaluation of eight open-source and two closed-source VideoLLMs on the proposed EventHallusion benchmark, we observe that the open-source models suffer significantly from hallucination problems, whereas the closed-source ones perform markedly better. By further equipping open-source VideoLLMs with the proposed TCD approach, evident performance improvements are achieved across most metrics in the EventHallusion benchmark. Our codes and benchmark data are available at https://github.com/Stevetich/EventHallusion.
♻ ☆ Fast, Scale-Adaptive, and Uncertainty-Aware Downscaling of Earth System Model Fields with Generative Machine Learning
Accurate and high-resolution Earth system model (ESM) simulations are essential to assess the ecological and socio-economic impacts of anthropogenic climate change, but are computationally too expensive to be run at sufficiently high spatial resolution. Recent machine learning approaches have shown promising results in downscaling ESM simulations, outperforming state-of-the-art statistical approaches. However, existing methods require computationally costly retraining for each ESM and extrapolate poorly to climates unseen during training. We address these shortcomings by learning a consistency model (CM) that efficiently and accurately downscales arbitrary ESM simulations without retraining in a zero-shot manner. Our approach yields probabilistic downscaled fields at a resolution only limited by the observational reference data. We show that the CM outperforms state-of-the-art diffusion models at a fraction of computational cost while maintaining high controllability on the downscaling task. Further, our method generalizes to climate states unseen during training without explicitly formulated physical constraints.
♻ ☆ Learning Symmetries via Weight-Sharing with Doubly Stochastic Tensors
Group equivariance has emerged as a valuable inductive bias in deep learning, enhancing generalization, data efficiency, and robustness. Classically, group equivariant methods require the groups of interest to be known beforehand, which may not be realistic for real-world data. Additionally, baking in fixed group equivariance may impose overly restrictive constraints on model architecture. This highlights the need for methods that can dynamically discover and apply symmetries as soft constraints. For neural network architectures, equivariance is commonly achieved through group transformations of a canonical weight tensor, resulting in weight sharing over a given group $G$. In this work, we propose to learn such a weight-sharing scheme by defining a collection of learnable doubly stochastic matrices that act as soft permutation matrices on canonical weight tensors, which can take regular group representations as a special case. This yields learnable kernel transformations that are jointly optimized with downstream tasks. We show that when the dataset exhibits strong symmetries, the permutation matrices will converge to regular group representations and our weight-sharing networks effectively become regular group convolutions. Additionally, the flexibility of the method enables it to effectively pick up on partial symmetries.
comment: 19 pages, 14 figures, 4 tables
♻ ☆ TextureCrop: Enhancing Synthetic Image Detection through Texture-based Cropping
Generative AI technologies produce increasingly realistic imagery, which, despite its potential for creative applications, can also be misused to produce misleading and harmful content. This renders Synthetic Image Detection (SID) methods essential for identifying AI-generated content online. State-of-the-art SID methods typically resize or center-crop input images due to architectural or computational constraints, which hampers the detection of artifacts that appear in high-resolution images. To address this limitation, we propose TextureCrop, an image pre-processing component that can be plugged in any pre-trained SID model to improve its performance. By focusing on high-frequency image parts where generative artifacts are prevalent, TextureCrop enhances SID performance with manageable memory requirements. Experimental results demonstrate a consistent improvement in AUC across various detectors by 6.1% compared to center cropping and by 15% compared to resizing, across high-resolution images from the Forensynths, Synthbuster and TWIGMA datasets. Code available at https : //github.com/mever-team/texture-crop.
comment: 10 pages, 7 images
♻ ☆ Transformers and Large Language Models for Efficient Intrusion Detection Systems: A Comprehensive Survey
With significant advancements in Transformers LLMs, NLP has extended its reach into many research fields due to its enhanced capabilities in text generation and user interaction. One field benefiting greatly from these advancements is cybersecurity. In cybersecurity, many parameters that need to be protected and exchanged between senders and receivers are in the form of text and tabular data, making NLP a valuable tool in enhancing the security measures of communication protocols. This survey paper provides a comprehensive analysis of the utilization of Transformers and LLMs in cyber-threat detection systems. The methodology of paper selection and bibliometric analysis is outlined to establish a rigorous framework for evaluating existing research. The fundamentals of Transformers are discussed, including background information on various cyber-attacks and datasets commonly used in this field. The survey explores the application of Transformers in IDSs, focusing on different architectures such as Attention-based models, LLMs like BERT and GPT, CNN/LSTM-Transformer hybrids, emerging approaches like ViTs, among others. Furthermore, it explores the diverse environments and applications where Transformers and LLMs-based IDS have been implemented, including computer networks, IoT devices, critical infrastructure protection, cloud computing, SDN, as well as in autonomous vehicles. The paper also addresses research challenges and future directions in this area, identifying key issues such as interpretability, scalability, and adaptability to evolving threats, and more. Finally, the conclusion summarizes the findings and highlights the significance of Transformers and LLMs in enhancing cyber-threat detection capabilities, while also outlining potential avenues for further research and development.
comment: arXiv admin note: text overlap with arXiv:2405.04760 by other authors
♻ ☆ Rethinking Decoders for Transformer-based Semantic Segmentation: A Compression Perspective NeurIPS2024
State-of-the-art methods for Transformer-based semantic segmentation typically adopt Transformer decoders that are used to extract additional embeddings from image embeddings via cross-attention, refine either or both types of embeddings via self-attention, and project image embeddings onto the additional embeddings via dot-product. Despite their remarkable success, these empirical designs still lack theoretical justifications or interpretations, thus hindering potentially principled improvements. In this paper, we argue that there are fundamental connections between semantic segmentation and compression, especially between the Transformer decoders and Principal Component Analysis (PCA). From such a perspective, we derive a white-box, fully attentional DEcoder for PrIncipled semantiC segemenTation (DEPICT), with the interpretations as follows: 1) the self-attention operator refines image embeddings to construct an ideal principal subspace that aligns with the supervision and retains most information; 2) the cross-attention operator seeks to find a low-rank approximation of the refined image embeddings, which is expected to be a set of orthonormal bases of the principal subspace and corresponds to the predefined classes; 3) the dot-product operation yields compact representation for image embeddings as segmentation masks. Experiments conducted on dataset ADE20K find that DEPICT consistently outperforms its black-box counterpart, Segmenter, and it is light weight and more robust.
comment: NeurIPS2024. Code:https://github.com/QishuaiWen/DEPICT/
♻ ☆ Enhanced Masked Image Modeling to Avoid Model Collapse on Multi-modal MRI Datasets
Multi-modal magnetic resonance imaging (MRI) provides information of lesions for computer-aided diagnosis from different views. Deep learning algorithms are suitable for identifying specific anatomical structures, segmenting lesions, and classifying diseases. Manual labels are limited due to the high expense, which hinders further improvement of accuracy. Self-supervised learning, particularly masked image modeling (MIM), has shown promise in utilizing unlabeled data. However, we spot model collapse when applying MIM to multi-modal MRI datasets. The performance of downstream tasks does not see any improvement following the collapsed model. To solve model collapse, we analyze and address it in two types: complete collapse and dimensional collapse. We find complete collapse occurs because the collapsed loss value in multi-modal MRI datasets falls below the normally converged loss value. Based on this, the hybrid mask pattern (HMP) masking strategy is introduced to elevate the collapsed loss above the normally converged loss value and avoid complete collapse. Additionally, we reveal that dimensional collapse stems from insufficient feature uniformity in MIM. We mitigate dimensional collapse by introducing the pyramid barlow twins (PBT) module as an explicit regularization method. Overall, we construct the enhanced MIM (E-MIM) with HMP and PBT module to avoid model collapse multi-modal MRI. Experiments are conducted on three multi-modal MRI datasets to validate the effectiveness of our approach in preventing both types of model collapse. By preventing model collapse, the training of the model becomes more stable, resulting in a decent improvement in performance for segmentation and classification tasks. The code is available at https://github.com/LinxuanHan/E-MIM.
♻ ☆ Perception Matters: Enhancing Embodied AI with Uncertainty-Aware Semantic Segmentation
Embodied AI has made significant progress acting in unexplored environments. However, tasks such as object search have largely focused on efficient policy learning. In this work, we identify several gaps in current search methods: They largely focus on dated perception models, neglect temporal aggregation, and transfer from ground truth directly to noisy perception at test time, without accounting for the resulting overconfidence in the perceived state. We address the identified problems through calibrated perception probabilities and uncertainty across aggregation and found decisions, thereby adapting the models for sequential tasks. The resulting methods can be directly integrated with pretrained models across a wide family of existing search approaches at no additional training cost. We perform extensive evaluations of aggregation methods across both different semantic perception models and policies, confirming the importance of calibrated uncertainties in both the aggregation and found decisions. We make the code and trained models available at https://semantic-search.cs.uni-freiburg.de.
♻ ☆ TextureDiffusion: Target Prompt Disentangled Editing for Various Texture Transfer
Recently, text-guided image editing has achieved significant success. However, existing methods can only apply simple textures like wood or gold when changing the texture of an object. Complex textures such as cloud or fire pose a challenge. This limitation stems from that the target prompt needs to contain both the input image content and , restricting the texture representation. In this paper, we propose TextureDiffusion, a tuning-free image editing method applied to various texture transfer. Initially, the target prompt is directly set to "", making the texture disentangled from the input image content to enhance texture representation. Subsequently, query features in self-attention and features in residual blocks are utilized to preserve the structure of the input image. Finally, to maintain the background, we introduce an edit localization technique which blends the self-attention results and the intermediate latents. Comprehensive experiments demonstrate that TextureDiffusion can harmoniously transfer various textures with excellent structure and background preservation. Code is publicly available at https://github.com/THU-CVML/TextureDiffusion
♻ ☆ ONER: Online Experience Replay for Incremental Anomaly Detection
Incremental anomaly detection sequentially recognizes abnormal regions in novel categories for dynamic industrial scenarios. This remains highly challenging due to knowledge overwriting and feature conflicts, leading to catastrophic forgetting. In this work, we propose ONER, an end-to-end ONline Experience Replay method, which efficiently mitigates catastrophic forgetting while adapting to new tasks with minimal cost. Specifically, our framework utilizes two types of experiences from past tasks: decomposed prompts and semantic prototypes, addressing both model parameter updates and feature optimization. The decomposed prompts consist of learnable components that assemble to produce attention-conditioned prompts. These prompts reuse previously learned knowledge, enabling model to learn novel tasks effectively. The semantic prototypes operate at both pixel and image levels, performing regularization in the latent feature space to prevent forgetting across various tasks. Extensive experiments demonstrate that our method achieves state-of-the-art performance in incremental anomaly detection with significantly reduced forgetting, as well as efficiently adapting to new categories with minimal costs. These results confirm the efficiency and stability of ONER, making it a powerful solution for real-world applications.
♻ ☆ HyFusion: Enhanced Reception Field Transformer for Hyperspectral Image Fusion
Hyperspectral image (HSI) fusion addresses the challenge of reconstructing High-Resolution HSIs (HR-HSIs) from High-Resolution Multispectral images (HR-MSIs) and Low-Resolution HSIs (LR-HSIs), a critical task given the high costs and hardware limitations associated with acquiring high-quality HSIs. While existing methods leverage spatial and spectral relationships, they often suffer from limited receptive fields and insufficient feature utilization, leading to suboptimal performance. Furthermore, the scarcity of high-quality HSI data highlights the importance of efficient data utilization to maximize reconstruction quality. To address these issues, we propose HyFusion, a novel Dual-Coupled Network (DCN) framework designed to enhance cross-domain feature extraction and enable effective feature map reusing. The framework first processes HR-MSI and LR-HSI inputs through specialized subnetworks that mutually enhance each other during feature extraction, preserving complementary spatial and spectral details. At its core, HyFusion utilizes an Enhanced Reception Field Block (ERFB), which combines shifting-window attention and dense connections to expand the receptive field, effectively capturing long-range dependencies while minimizing information loss. Extensive experiments demonstrate that HyFusion achieves state-of-the-art performance in HR-MSI/LR-HSI fusion, significantly improving reconstruction quality while maintaining a compact model size and computational efficiency. By integrating enhanced receptive fields and feature map reusing into a coupled network architecture, HyFusion provides a practical and effective solution for HSI fusion in resource-constrained scenarios, setting a new benchmark in hyperspectral imaging. Our code will be publicly available.
comment: Submitted to IGARSS 2025
♻ ☆ Knowledge-Guided Prompt Learning for Deepfake Facial Image Detection ICASSP 2025
Recent generative models demonstrate impressive performance on synthesizing photographic images, which makes humans hardly to distinguish them from pristine ones, especially on realistic-looking synthetic facial images. Previous works mostly focus on mining discriminative artifacts from vast amount of visual data. However, they usually lack the exploration of prior knowledge and rarely pay attention to the domain shift between training categories (e.g., natural and indoor objects) and testing ones (e.g., fine-grained human facial images), resulting in unsatisfactory detection performance. To address these issues, we propose a novel knowledge-guided prompt learning method for deepfake facial image detection. Specifically, we retrieve forgery-related prompts from large language models as expert knowledge to guide the optimization of learnable prompts. Besides, we elaborate test-time prompt tuning to alleviate the domain shift, achieving significant performance improvement and boosting the application in real-world scenarios. Extensive experiments on DeepFakeFaceForensics dataset show that our proposed approach notably outperforms state-of-the-art methods.
comment: Accepted by ICASSP 2025
♻ ☆ PastNet: Introducing Physical Inductive Biases for Spatio-temporal Video Prediction
In this paper, we investigate the challenge of spatio-temporal video prediction task, which involves generating future video frames based on historical spatio-temporal observation streams. Existing approaches typically utilize external information such as semantic maps to improve video prediction accuracy, which often neglect the inherent physical knowledge embedded within videos. Worse still, their high computational costs could impede their applications for high-resolution videos. To address these constraints, we introduce a novel framework called \underline{P}hysics-\underline{a}ssisted \underline{S}patio-\underline{t}emporal \underline{Net}work (PastNet) for high-quality video prediction. The core of PastNet lies in incorporating a spectral convolution operator in the Fourier domain, which efficiently introduces inductive biases from the underlying physical laws. Additionally, we employ a memory bank with the estimated intrinsic dimensionality to discretize local features during the processing of complex spatio-temporal signals, thereby reducing computational costs and facilitating efficient high-resolution video prediction. Extensive experiments on various widely-used spatio-temporal video benchmarks demonstrate the effectiveness and efficiency of the proposed PastNet compared with a range of state-of-the-art methods, particularly in high-resolution scenarios.
comment: 11
♻ ☆ DehazeGS: Seeing Through Fog with 3D Gaussian Splatting
Current novel view synthesis tasks primarily rely on high-quality and clear images. However, in foggy scenes, scattering and attenuation can significantly degrade the reconstruction and rendering quality. Although NeRF-based dehazing reconstruction algorithms have been developed, their use of deep fully connected neural networks and per-ray sampling strategies leads to high computational costs. Moreover, NeRF's implicit representation struggles to recover fine details from hazy scenes. In contrast, recent advancements in 3D Gaussian Splatting achieve high-quality 3D scene reconstruction by explicitly modeling point clouds into 3D Gaussians. In this paper, we propose leveraging the explicit Gaussian representation to explain the foggy image formation process through a physically accurate forward rendering process. We introduce DehazeGS, a method capable of decomposing and rendering a fog-free background from participating media using only muti-view foggy images as input. We model the transmission within each Gaussian distribution to simulate the formation of fog. During this process, we jointly learn the atmospheric light and scattering coefficient while optimizing the Gaussian representation of the hazy scene. In the inference stage, we eliminate the effects of scattering and attenuation on the Gaussians and directly project them onto a 2D plane to obtain a clear view. Experiments on both synthetic and real-world foggy datasets demonstrate that DehazeGS achieves state-of-the-art performance in terms of both rendering quality and computational efficiency.
comment: 9 pages,4 figures
♻ ☆ Spacewalker: Traversing Representation Spaces for Fast Interactive Exploration and Annotation of Unstructured Data
In industries such as healthcare, finance, and manufacturing, analysis of unstructured textual data presents significant challenges for analysis and decision making. Uncovering patterns within large-scale corpora and understanding their semantic impact is critical, but depends on domain experts or resource-intensive manual reviews. In response, we introduce Spacewalker in this system demonstration paper, an interactive tool designed to analyze, explore, and annotate data across multiple modalities. It allows users to extract data representations, visualize them in low-dimensional spaces and traverse large datasets either exploratory or by querying regions of interest. We evaluated Spacewalker through extensive experiments and annotation studies, assessing its efficacy in improving data integrity verification and annotation. We show that Spacewalker reduces time and effort compared to traditional methods. The code of this work is open-source and can be found at: https://github.com/code-lukas/Spacewalker
♻ ☆ Knowledge Transfer and Domain Adaptation for Fine-Grained Remote Sensing Image Segmentation
Fine-grained remote sensing image segmentation is essential for accurately identifying detailed objects in remote sensing images. Recently, vision transformer models (VTMs) pre-trained on large-scale datasets have demonstrated strong zero-shot generalization. However, directly applying them to specific tasks may lead to domain shift. We introduce a novel end-to-end learning paradigm combining knowledge guidance with domain refinement to enhance performance. We present two key components: the Feature Alignment Module (FAM) and the Feature Modulation Module (FMM). FAM aligns features from a CNN-based backbone with those from the pretrained VTM's encoder using channel transformation and spatial interpolation, and transfers knowledge via KL divergence and L2 normalization constraint. FMM further adapts the knowledge to the specific domain to address domain shift. We also introduce a fine-grained grass segmentation dataset and demonstrate, through experiments on two datasets, that our method achieves a significant improvement of 2.57 mIoU on the grass dataset and 3.73 mIoU on the cloud dataset. The results highlight the potential of combining knowledge transfer and domain adaptation to overcome domain-related challenges and data limitations. The project page is available at https://xavierjiezou.github.io/KTDA/.
comment: 6 pages, 3 figures, 6 tables
♻ ☆ Edicho: Consistent Image Editing in the Wild
As a verified need, consistent editing across in-the-wild images remains a technical challenge arising from various unmanageable factors, like object poses, lighting conditions, and photography environments. Edicho steps in with a training-free solution based on diffusion models, featuring a fundamental design principle of using explicit image correspondence to direct editing. Specifically, the key components include an attention manipulation module and a carefully refined classifier-free guidance (CFG) denoising strategy, both of which take into account the pre-estimated correspondence. Such an inference-time algorithm enjoys a plug-and-play nature and is compatible to most diffusion-based editing methods, such as ControlNet and BrushNet. Extensive results demonstrate the efficacy of Edicho in consistent cross-image editing under diverse settings. We will release the code to facilitate future studies.
comment: Project page: https://ant-research.github.io/edicho/
♻ ☆ MoPE: Mixture of Prompt Experts for Parameter-Efficient and Scalable Multimodal Fusion
Despite the demonstrated parameter efficiency of prompt-based multimodal fusion methods, their limited adaptivity and expressiveness often result in suboptimal performance compared to other tuning approaches. In this paper, we introduce the Mixture of Prompt Experts (MoPE), the first technique designed to overcome these limitations by decomposing standard prompts to capture instance-level features adaptively. Building on this decomposition, MoPE enhances prompt fusion's expressiveness by leveraging multimodal pairing priors to route the most effective prompt for each instance dynamically. Compared to vanilla prompting, our MoPE-based fusion method exhibits greater expressiveness, scaling more effectively with the training data and the overall number of trainable parameters. We also investigate regularization terms for expert routing, which lead to emergent expert specialization with enhanced adaptiveness and interpretablity. Extensive experiments across six multimodal datasets spanning four modalities demonstrate state-of-the-art performance for prompt fusion, matching or even surpassing the performance of fine-tuning while requiring only 0.8% of the trainable parameters. Project homepage: https://github.com/songrise/MoPE
comment: Under Review, Extended version of arxiv:2312.03734
♻ ☆ BIOMEDICA: An Open Biomedical Image-Caption Archive, Dataset, and Vision-Language Models Derived from Scientific Literature
The development of vision-language models (VLMs) is driven by large-scale and diverse multimodal datasets. However, progress toward generalist biomedical VLMs is limited by the lack of annotated, publicly accessible datasets across biology and medicine. Existing efforts are restricted to narrow domains, missing the full diversity of biomedical knowledge encoded in scientific literature. To address this gap, we introduce BIOMEDICA, a scalable, open-source framework to extract, annotate, and serialize the entirety of the PubMed Central Open Access subset into an easy-to-use, publicly accessible dataset. Our framework produces a comprehensive archive with over 24 million unique image-text pairs from over 6 million articles. Metadata and expert-guided annotations are also provided. We demonstrate the utility and accessibility of our resource by releasing BMCA-CLIP, a suite of CLIP-style models continuously pre-trained on the BIOMEDICA dataset via streaming, eliminating the need to download 27 TB of data locally. On average, our models achieve state-of-the-art performance across 40 tasks - spanning pathology, radiology, ophthalmology, dermatology, surgery, molecular biology, parasitology, and cell biology - excelling in zero-shot classification with a 6.56% average improvement (as high as 29.8% and 17.5% in dermatology and ophthalmology, respectively), and stronger image-text retrieval, all while using 10x less compute. To foster reproducibility and collaboration, we release our codebase and dataset for the broader research community.
♻ ☆ Recognizing Artistic Style of Archaeological Image Fragments Using Deep Style Extrapolation
Ancient artworks obtained in archaeological excavations usually suffer from a certain degree of fragmentation and physical degradation. Often, fragments of multiple artifacts from different periods or artistic styles could be found on the same site. With each fragment containing only partial information about its source, and pieces from different objects being mixed, categorizing broken artifacts based on their visual cues could be a challenging task, even for professionals. As classification is a common function of many machine learning models, the power of modern architectures can be harnessed for efficient and accurate fragment classification. In this work, we present a generalized deep-learning framework for predicting the artistic style of image fragments, achieving state-of-the-art results for pieces with varying styles and geometries.
comment: To be published in the 27th International Conference on Human-Computer Interaction (HCII 2025)
♻ ☆ Flash Window Attention: speedup the attention computation for Swin Transformer
To address the high resolution of image pixels, the Swin Transformer introduces window attention. This mechanism divides an image into non-overlapping windows and restricts attention computation to within each window, significantly enhancing computational efficiency. To further optimize this process, one might consider replacing standard attention with flash attention, which has proven to be more efficient in language models. However, a direct substitution is ineffective. Flash attention is designed for long sequences, whereas window attention deals with shorter sequences but must handle numerous of them in parallel. In this report, we present an optimized solution called Flash Window Attention, tailored specifically for window attention. Flash Window Attention improves attention computation efficiency by up to 300% and enhances end-to-end runtime efficiency by up to 30%. Our code is available online.
♻ ☆ Analyzing Infrastructure LiDAR Placement with Realistic LiDAR Simulation Library ICRA'23
Recently, Vehicle-to-Everything(V2X) cooperative perception has attracted increasing attention. Infrastructure sensors play a critical role in this research field; however, how to find the optimal placement of infrastructure sensors is rarely studied. In this paper, we investigate the problem of infrastructure sensor placement and propose a pipeline that can efficiently and effectively find optimal installation positions for infrastructure sensors in a realistic simulated environment. To better simulate and evaluate LiDAR placement, we establish a Realistic LiDAR Simulation library that can simulate the unique characteristics of different popular LiDARs and produce high-fidelity LiDAR point clouds in the CARLA simulator. Through simulating point cloud data in different LiDAR placements, we can evaluate the perception accuracy of these placements using multiple detection models. Then, we analyze the correlation between the point cloud distribution and perception accuracy by calculating the density and uniformity of regions of interest. Experiments show that when using the same number and type of LiDAR, the placement scheme optimized by our proposed method improves the average precision by 15%, compared with the conventional placement scheme in the standard lane scene. We also analyze the correlation between perception performance in the region of interest and LiDAR point cloud distribution and validate that density and uniformity can be indicators of performance. Both the RLS Library and related code will be released at https://github.com/PJLab-ADG/PCSim.
comment: 7 pages, 6 figures, accepted to the IEEE International Conference on Robotics and Automation (ICRA'23)
♻ ☆ A Cascaded Dilated Convolution Approach for Mpox Lesion Classification
The global outbreak of the Mpox virus, classified as a Public Health Emergency of International Concern (PHEIC) by the World Health Organization, presents significant diagnostic challenges due to its visual similarity to other skin lesion diseases. Traditional diagnostic methods for Mpox, which rely on clinical symptoms and laboratory tests, are slow and labor intensive. Deep learning-based approaches for skin lesion classification offer a promising alternative. However, developing a model that balances efficiency with accuracy is crucial to ensure reliable and timely diagnosis without compromising performance. This study introduces the Cascaded Atrous Group Attention (CAGA) framework to address these challenges, combining the Cascaded Atrous Attention module and the Cascaded Group Attention mechanism. The Cascaded Atrous Attention module utilizes dilated convolutions and cascades the outputs to enhance multi-scale representation. This is integrated into the Cascaded Group Attention mechanism, which reduces redundancy in Multi-Head Self-Attention. By integrating the Cascaded Atrous Group Attention module with EfficientViT-L1 as the backbone architecture, this approach achieves state-of-the-art performance, reaching an accuracy of 98% on the Mpox Close Skin Image (MCSI) dataset while reducing model parameters by 37.5% compared to the original EfficientViT-L1. The model's robustness is demonstrated through extensive validation on two additional benchmark datasets, where it consistently outperforms existing approaches.
comment: 8 pages, 4 figures, Submitted to Medical Imaging with Deep Learning
♻ ☆ Implicit Neural Representations with Fourier Kolmogorov-Arnold Networks ICASSP 2025
Implicit neural representations (INRs) use neural networks to provide continuous and resolution-independent representations of complex signals with a small number of parameters. However, existing INR models often fail to capture important frequency components specific to each task. To address this issue, in this paper, we propose a Fourier Kolmogorov Arnold network (FKAN) for INRs. The proposed FKAN utilizes learnable activation functions modeled as Fourier series in the first layer to effectively control and learn the task-specific frequency components. In addition, the activation functions with learnable Fourier coefficients improve the ability of the network to capture complex patterns and details, which is beneficial for high-resolution and high-dimensional data. Experimental results show that our proposed FKAN model outperforms three state-of-the-art baseline schemes, and improves the peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) for the image representation task and intersection over union (IoU) for the 3D occupancy volume representation task, respectively. The code is available at github.com/Ali-Meh619/FKAN.
comment: Accepted for publication in Proc. IEEE ICASSP 2025
♻ ☆ Gradient descent with generalized Newton's method
We propose the generalized Newton's method (GeN) -- a Hessian-informed approach that applies to any optimizer such as SGD and Adam, and covers the Newton-Raphson method as a sub-case. Our method automatically and dynamically selects the learning rate that accelerates the convergence, without the intensive tuning of the learning rate scheduler. In practice, our method is easily implementable, since it only requires additional forward passes with almost zero computational overhead (in terms of training time and memory cost), if the overhead is amortized over many iterations. We present extensive experiments on language and vision tasks (e.g. GPT and ResNet) to showcase that GeN optimizers match the state-of-the-art performance, which was achieved with carefully tuned learning rate schedulers.
♻ ☆ MambaTrack: Exploiting Dual-Enhancement for Night UAV Tracking
Night unmanned aerial vehicle (UAV) tracking is impeded by the challenges of poor illumination, with previous daylight-optimized methods demonstrating suboptimal performance in low-light conditions, limiting the utility of UAV applications. To this end, we propose an efficient mamba-based tracker, leveraging dual enhancement techniques to boost night UAV tracking. The mamba-based low-light enhancer, equipped with an illumination estimator and a damage restorer, achieves global image enhancement while preserving the details and structure of low-light images. Additionally, we advance a cross-modal mamba network to achieve efficient interactive learning between vision and language modalities. Extensive experiments showcase that our method achieves advanced performance and exhibits significantly improved computation and memory efficiency. For instance, our method is 2.8$\times$ faster than CiteTracker and reduces 50.2$\%$ GPU memory. Our codes are available at \url{https://github.com/983632847/Awesome-Multimodal-Object-Tracking}.
comment: Preprint
♻ ☆ Dissecting Query-Key Interaction in Vision Transformers
Self-attention in vision transformers is often thought to perform perceptual grouping where tokens attend to other tokens with similar embeddings, which could correspond to semantically similar features of an object. However, attending to dissimilar tokens can be beneficial by providing contextual information. We propose to analyze the query-key interaction by the singular value decomposition of the interaction matrix (i.e. ${\textbf{W}_q}^\top\textbf{W}_k$). We find that in many ViTs, especially those with classification training objectives, early layers attend more to similar tokens, while late layers show increased attention to dissimilar tokens, providing evidence corresponding to perceptual grouping and contextualization, respectively. Many of these interactions between features represented by singular vectors are interpretable and semantic, such as attention between relevant objects, between parts of an object, or between the foreground and background. This offers a novel perspective on interpreting the attention mechanism, which contributes to understanding how transformer models utilize context and salient features when processing images.
♻ ☆ Smartphone-based Eye Tracking System using Edge Intelligence and Model Optimisation
A significant limitation of current smartphone-based eye-tracking algorithms is their low accuracy when applied to video-type visual stimuli, as they are typically trained on static images. Also, the increasing demand for real-time interactive applications like games, VR, and AR on smartphones requires overcoming the limitations posed by resource constraints such as limited computational power, battery life, and network bandwidth. Therefore, we developed two new smartphone eye-tracking techniques for video-type visuals by combining Convolutional Neural Networks (CNN) with two different Recurrent Neural Networks (RNN), namely Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU). Our CNN+LSTM and CNN+GRU models achieved an average Root Mean Square Error of 0.955 cm and 1.091 cm, respectively. To address the computational constraints of smartphones, we developed an edge intelligence architecture to enhance the performance of smartphone-based eye tracking. We applied various optimisation methods like quantisation and pruning to deep learning models for better energy, CPU, and memory usage on edge devices, focusing on real-time processing. Using model quantisation, the model inference time in the CNN+LSTM and CNN+GRU models was reduced by 21.72% and 19.50%, respectively, on edge devices.
comment: I have included the three papers as reference, which are closely related. We have expanded the future work section to provide a more thorough discussion of the concepts of "varying lighting conditions" and "dynamic user environments." We have added a note below Table 4 to clarify the abbreviations' meaning. Elaborated the role of the Domain Expert within the presentation layer in Section 4.1
♻ ☆ The Collection of a Human Robot Collaboration Dataset for Cooperative Assembly in Glovebox Environments
Industry 4.0 introduced AI as a transformative solution for modernizing manufacturing processes. Its successor, Industry 5.0, envisions humans as collaborators and experts guiding these AI-driven manufacturing solutions. Developing these techniques necessitates algorithms capable of safe, real-time identification of human positions in a scene, particularly their hands, during collaborative assembly. Although substantial efforts have curated datasets for hand segmentation, most focus on residential or commercial domains. Existing datasets targeting industrial settings predominantly rely on synthetic data, which we demonstrate does not effectively transfer to real-world operations. Moreover, these datasets lack uncertainty estimations critical for safe collaboration. Addressing these gaps, we present HAGS: Hand and Glove Segmentation Dataset. This dataset provides challenging examples to build applications toward hand and glove segmentation in industrial human-robot collaboration scenarios as well as assess out-of-distribution images, constructed via green screen augmentations, to determine ML-classifier robustness. We study state-of-the-art, real-time segmentation models to evaluate existing methods. Our dataset and baselines are publicly available.
comment: draft paper to be submitted to IJRR
♻ ☆ A systematic review of the use of Deep Learning in Satellite Imagery for Agriculture
Agricultural research is essential for increasing food production to meet the requirements of an increasing population in the coming decades. Recently, satellite technology has been improving rapidly and deep learning has seen much success in generic computer vision tasks and many application areas which presents an important opportunity to improve analysis of agricultural land. Here we present a systematic review of 150 studies to find the current uses of deep learning on satellite imagery for agricultural research. Although we identify 5 categories of agricultural monitoring tasks, the majority of the research interest is in crop segmentation and yield prediction. We found that, when used, modern deep learning methods consistently outperformed traditional machine learning across most tasks; the only exception was that Long Short-Term Memory (LSTM) Recurrent Neural Networks did not consistently outperform Random Forests (RF) for yield prediction. The reviewed studies have largely adopted methodologies from generic computer vision, except for one major omission: benchmark datasets are not utilised to evaluate models across studies, making it difficult to compare results. Additionally, some studies have specifically utilised the extra spectral resolution available in satellite imagery, but other divergent properties of satellite images - such as the hugely different scales of spatial patterns - are not being taken advantage of in the reviewed studies.
comment: 23 pages, 5 figures and 10 tables in main paper. Final version, as submitted and accepted at JSTARS
♻ ☆ Zero-shot 3D Segmentation of Abdominal Organs in CT Scans Using Segment Anything Model 2: Adapting Video Tracking Capabilities for 3D Medical Imaging
Objectives: To evaluate the zero-shot performance of Segment Anything Model 2 (SAM 2) in 3D segmentation of abdominal organs in CT scans, and to investigate the effects of prompt settings on segmentation results. Materials and Methods: In this retrospective study, we used a subset of the TotalSegmentator CT dataset from eight institutions to assess SAM 2's ability to segment eight abdominal organs. Segmentation was initiated from three different z-coordinate levels (caudal, mid, and cranial levels) of each organ. Performance was measured using the Dice similarity coefficient (DSC). We also analyzed the impact of "negative prompts," which explicitly exclude certain regions from the segmentation process, on accuracy. Results: 123 patients (mean age, 60.7 \pm 15.5 years; 63 men, 60 women) were evaluated. As a zero-shot approach, larger organs with clear boundaries demonstrated high segmentation performance, with mean DSCs as follows: liver 0.821 \pm 0.192, right kidney 0.862 \pm 0.212, left kidney 0.870 \pm 0.154, and spleen 0.891 \pm 0.131. Smaller organs showed lower performance: gallbladder 0.531 \pm 0.291, pancreas 0.361 \pm 0.197, and adrenal glands, right 0.203 \pm 0.222, left 0.308 \pm 0.234. The initial slice for segmentation and the use of negative prompts significantly influenced the results. By removing negative prompts from the input, the DSCs significantly decreased for six organs. Conclusion: SAM 2 demonstrated promising zero-shot performance in segmenting certain abdominal organs in CT scans, particularly larger organs. Performance was significantly influenced by input negative prompts and initial slice selection, highlighting the importance of optimizing these factors.
comment: 20 pages, 7 figures (including 2 supplemental figure), 4 tables
♻ ☆ XVertNet: Unsupervised Contrast Enhancement of Vertebral Structures with Dynamic Self-Tuning Guidance and Multi-Stage Analysis
Chest X-rays remain the primary diagnostic tool in emergency medicine, yet their limited ability to capture fine anatomical details can result in missed or delayed diagnoses. To address this, we introduce XVertNet, a novel deep-learning framework designed to enhance vertebral structure visualization in X-ray images significantly. Our framework introduces two key innovations: (1) An unsupervised learning architecture that eliminates reliance on manually labeled training data a persistent bottleneck in medical imaging, and (2) a dynamic self-tuned internal guidance mechanism featuring an adaptive feedback loop for real-time image optimization. Extensive validation across four major public datasets revealed that XVertNet outperforms state-of-the-art enhancement methods, as demonstrated by improvements in entropy scores, Tenengrad criterion values, the local phase coherence sharpness index (LPC-SI), and thetone mapped image quality index (TMQI). Furthermore, clinical validation conducted with two board-certified radiologists confirmed that the enhanced images enabled more sensitive detection of subtle vertebral fractures and degenerative changes. The unsupervised nature of XVertNet facilitates immediate clinical deployment without requiring additional training overhead. This innovation represents a transformative advancement in emergency radiology, providing a scalable and time-efficient solution to enhance diagnostic accuracy in high-pressure clinical environments.
comment: 13 pages
♻ ☆ Expressive Text-to-Image Generation with Rich Text
Plain text has become a prevalent interface for text-to-image synthesis. However, its limited customization options hinder users from accurately describing desired outputs. For example, plain text makes it hard to specify continuous quantities, such as the precise RGB color value or importance of each word. Furthermore, creating detailed text prompts for complex scenes is tedious for humans to write and challenging for text encoders to interpret. To address these challenges, we propose using a rich-text editor supporting formats such as font style, size, color, and footnote. We extract each word's attributes from rich text to enable local style control, explicit token reweighting, precise color rendering, and detailed region synthesis. We achieve these capabilities through a region-based diffusion process. We first obtain each word's region based on attention maps of a diffusion process using plain text. For each region, we enforce its text attributes by creating region-specific detailed prompts and applying region-specific guidance, and maintain its fidelity against plain-text generation through region-based injections. We present various examples of image generation from rich text and demonstrate that our method outperforms strong baselines with quantitative evaluations.
comment: Project webpage: https://rich-text-to-image.github.io/
♻ ☆ UrbanIR: Large-Scale Urban Scene Inverse Rendering from a Single Video
We present UrbanIR (Urban Scene Inverse Rendering), a new inverse graphics model that enables realistic, free-viewpoint renderings of scenes under various lighting conditions with a single video. It accurately infers shape, albedo, visibility, and sun and sky illumination from wide-baseline videos, such as those from car-mounted cameras, differing from NeRF's dense view settings. In this context, standard methods often yield subpar geometry and material estimates, such as inaccurate roof representations and numerous 'floaters'. UrbanIR addresses these issues with novel losses that reduce errors in inverse graphics inference and rendering artifacts. Its techniques allow for precise shadow volume estimation in the original scene. The model's outputs support controllable editing, enabling photorealistic free-viewpoint renderings of night simulations, relit scenes, and inserted objects, marking a significant improvement over existing state-of-the-art methods.
comment: https://urbaninverserendering.github.io/
♻ ☆ SYNAPSE: SYmbolic Neural-Aided Preference Synthesis Engine AAAI 25
This paper addresses the problem of preference learning, which aims to align robot behaviors through learning user specific preferences (e.g. "good pull-over location") from visual demonstrations. Despite its similarity to learning factual concepts (e.g. "red door"), preference learning is a fundamentally harder problem due to its subjective nature and the paucity of person-specific training data. We address this problem using a novel framework called SYNAPSE, which is a neuro-symbolic approach designed to efficiently learn preferential concepts from limited data. SYNAPSE represents preferences as neuro-symbolic programs, facilitating inspection of individual parts for alignment, in a domain-specific language (DSL) that operates over images and leverages a novel combination of visual parsing, large language models, and program synthesis to learn programs representing individual preferences. We perform extensive evaluations on various preferential concepts as well as user case studies demonstrating its ability to align well with dissimilar user preferences. Our method significantly outperforms baselines, especially when it comes to out of distribution generalization. We show the importance of the design choices in the framework through multiple ablation studies. Code, additional results, and supplementary material can be found on the website: https://amrl.cs.utexas.edu/synapse
comment: Accepted (oral) at AAAI 25
♻ ☆ Enhancing Performance of Point Cloud Completion Networks with Consistency Loss
Point cloud completion networks are conventionally trained to minimize the disparities between the completed point cloud and the ground-truth counterpart. However, an incomplete object-level point cloud can have multiple valid completion solutions when it is examined in isolation. This one-to-many mapping issue can cause contradictory supervision signals to the network because the loss function may produce different values for identical input-output pairs of the network. In many cases, this issue could adversely affect the network optimization process. In this work, we propose to enhance the conventional learning objective using a novel completion consistency loss to mitigate the one-to-many mapping problem. Specifically, the proposed consistency loss ensure that a point cloud completion network generates a coherent completion solution for incomplete objects originating from the same source point cloud. Experimental results across multiple well-established datasets and benchmarks demonstrated the proposed completion consistency loss have excellent capability to enhance the completion performance of various existing networks without any modification to the design of the networks. The proposed consistency loss enhances the performance of the point completion network without affecting the inference speed, thereby increasing the accuracy of point cloud completion. Notably, a state-of-the-art point completion network trained with the proposed consistency loss can achieve state-of-the-art accuracy on the challenging new MVP dataset. The code and result of experiment various point completion models using proposed consistency loss will be available at: https://github.com/kaist-avelab/ConsistencyLoss .
comment: First version of Paper "Enhancing Performance of Point Cloud Completion Networks with Consistency Loss" by Kevin Tirta Wijaya and Christofel Rio Goenawan. In process submission to Neurocomputing Journal 2024
♻ ☆ SplatMAP: Online Dense Monocular SLAM with 3D Gaussian Splatting
Achieving high-fidelity 3D reconstruction from monocular video remains challenging due to the inherent limitations of traditional methods like Structure-from-Motion (SfM) and monocular SLAM in accurately capturing scene details. While differentiable rendering techniques such as Neural Radiance Fields (NeRF) address some of these challenges, their high computational costs make them unsuitable for real-time applications. Additionally, existing 3D Gaussian Splatting (3DGS) methods often focus on photometric consistency, neglecting geometric accuracy and failing to exploit SLAM's dynamic depth and pose updates for scene refinement. We propose a framework integrating dense SLAM with 3DGS for real-time, high-fidelity dense reconstruction. Our approach introduces SLAM-Informed Adaptive Densification, which dynamically updates and densifies the Gaussian model by leveraging dense point clouds from SLAM. Additionally, we incorporate Geometry-Guided Optimization, which combines edge-aware geometric constraints and photometric consistency to jointly optimize the appearance and geometry of the 3DGS scene representation, enabling detailed and accurate SLAM mapping reconstruction. Experiments on the Replica and TUM-RGBD datasets demonstrate the effectiveness of our approach, achieving state-of-the-art results among monocular systems. Specifically, our method achieves a PSNR of 36.864, SSIM of 0.985, and LPIPS of 0.040 on Replica, representing improvements of 10.7%, 6.4%, and 49.4%, respectively, over the previous SOTA. On TUM-RGBD, our method outperforms the closest baseline by 10.2%, 6.6%, and 34.7% in the same metrics. These results highlight the potential of our framework in bridging the gap between photometric and geometric dense 3D scene representations, paving the way for practical and efficient monocular dense reconstruction.
♻ ☆ On the Geometry of Deep Learning
In this paper, we overview one promising avenue of progress at the mathematical foundation of deep learning: the connection between deep networks and function approximation by affine splines (continuous piecewise linear functions in multiple dimensions). In particular, we will overview work over the past decade on understanding certain geometrical properties of a deep network's affine spline mapping, in particular how it tessellates its input space. As we will see, the affine spline connection and geometrical viewpoint provide a powerful portal through which to view, analyze, and improve the inner workings of a deep network.
comment: Accepted for publication at 'Notices of the American Mathematical Society'
Computation and Language 88
☆ PokerBench: Training Large Language Models to become Professional Poker Players AAAI 2025
We introduce PokerBench - a benchmark for evaluating the poker-playing abilities of large language models (LLMs). As LLMs excel in traditional NLP tasks, their application to complex, strategic games like poker poses a new challenge. Poker, an incomplete information game, demands a multitude of skills such as mathematics, reasoning, planning, strategy, and a deep understanding of game theory and human psychology. This makes Poker the ideal next frontier for large language models. PokerBench consists of a comprehensive compilation of 11,000 most important scenarios, split between pre-flop and post-flop play, developed in collaboration with trained poker players. We evaluate prominent models including GPT-4, ChatGPT 3.5, and various Llama and Gemma series models, finding that all state-of-the-art LLMs underperform in playing optimal poker. However, after fine-tuning, these models show marked improvements. We validate PokerBench by having models with different scores compete with each other, demonstrating that higher scores on PokerBench lead to higher win rates in actual poker games. Through gameplay between our fine-tuned model and GPT-4, we also identify limitations of simple supervised fine-tuning for learning optimal playing strategy, suggesting the need for more advanced methodologies for effectively training language models to excel in games. PokerBench thus presents a unique benchmark for a quick and reliable evaluation of the poker-playing ability of LLMs as well as a comprehensive benchmark to study the progress of LLMs in complex game-playing scenarios. The dataset and code will be made available at: \url{https://github.com/pokerllm/pokerbench}.
comment: AAAI 2025
☆ Exploring Robustness of Multilingual LLMs on Real-World Noisy Data
Large Language Models (LLMs) are trained on Web data that might contain spelling errors made by humans. But do they become robust to similar real-world noise? In this paper, we investigate the effect of real-world spelling mistakes on the performance of 9 language models, with parameters ranging from 0.2B to 13B, in 3 different NLP tasks, namely Natural Language Inference (NLI), Name Entity Recognition (NER), and Intent Classification (IC). We perform our experiments on 6 different languages and build a dictionary of real-world noise for them using the Wikipedia edit history. We show that the performance gap of the studied models on the clean and noisy test data averaged across all the datasets and languages ranges from 2.3 to 4.3 absolute percentage points. In addition, mT5 models, in general, show more robustness compared to BLOOM, Falcon, and BERT-like models. In particular, mT5 (13B), was the most robust on average overall, across the 3 tasks, and in 4 of the 6 languages.
☆ Enhancing Automated Interpretability with Output-Centric Feature Descriptions
Automated interpretability pipelines generate natural language descriptions for the concepts represented by features in large language models (LLMs), such as plants or the first word in a sentence. These descriptions are derived using inputs that activate the feature, which may be a dimension or a direction in the model's representation space. However, identifying activating inputs is costly, and the mechanistic role of a feature in model behavior is determined both by how inputs cause a feature to activate and by how feature activation affects outputs. Using steering evaluations, we reveal that current pipelines provide descriptions that fail to capture the causal effect of the feature on outputs. To fix this, we propose efficient, output-centric methods for automatically generating feature descriptions. These methods use the tokens weighted higher after feature stimulation or the highest weight tokens after applying the vocabulary "unembedding" head directly to the feature. Our output-centric descriptions better capture the causal effect of a feature on model outputs than input-centric descriptions, but combining the two leads to the best performance on both input and output evaluations. Lastly, we show that output-centric descriptions can be used to find inputs that activate features previously thought to be "dead".
☆ MiniMax-01: Scaling Foundation Models with Lightning Attention
We introduce MiniMax-01 series, including MiniMax-Text-01 and MiniMax-VL-01, which are comparable to top-tier models while offering superior capabilities in processing longer contexts. The core lies in lightning attention and its efficient scaling. To maximize computational capacity, we integrate it with Mixture of Experts (MoE), creating a model with 32 experts and 456 billion total parameters, of which 45.9 billion are activated for each token. We develop an optimized parallel strategy and highly efficient computation-communication overlap techniques for MoE and lightning attention. This approach enables us to conduct efficient training and inference on models with hundreds of billions of parameters across contexts spanning millions of tokens. The context window of MiniMax-Text-01 can reach up to 1 million tokens during training and extrapolate to 4 million tokens during inference at an affordable cost. Our vision-language model, MiniMax-VL-01 is built through continued training with 512 billion vision-language tokens. Experiments on both standard and in-house benchmarks show that our models match the performance of state-of-the-art models like GPT-4o and Claude-3.5-Sonnet while offering 20-32 times longer context window. We publicly release MiniMax-01 at https://github.com/MiniMax-AI.
comment: A technical report from MiniMax. The authors are listed in alphabetical order. We open-sourced our MiniMax-01 at https://github.com/MiniMax-AI
☆ Everybody Likes to Sleep: A Computer-Assisted Comparison of Object Naming Data from 30 Languages
Object naming - the act of identifying an object with a word or a phrase - is a fundamental skill in interpersonal communication, relevant to many disciplines, such as psycholinguistics, cognitive linguistics, or language and vision research. Object naming datasets, which consist of concept lists with picture pairings, are used to gain insights into how humans access and select names for objects in their surroundings and to study the cognitive processes involved in converting visual stimuli into semantic concepts. Unfortunately, object naming datasets often lack transparency and have a highly idiosyncratic structure. Our study tries to make current object naming data transparent and comparable by using a multilingual, computer-assisted approach that links individual items of object naming lists to unified concepts. Our current sample links 17 object naming datasets that cover 30 languages from 10 different language families. We illustrate how the comparative dataset can be explored by searching for concepts that recur across the majority of datasets and comparing the conceptual spaces of covered object naming datasets with classical basic vocabulary lists from historical linguistics and linguistic typology. Our findings can serve as a basis for enhancing cross-linguistic object naming research and as a guideline for future studies dealing with object naming tasks.
comment: To appear in the Proceedings of the Global WordNet Conference 2025
☆ Quantifying the Importance of Data Alignment in Downstream Model Performance
Contrary to the conventional emphasis on dataset size, we explore the role of data alignment -- an often overlooked aspect of data quality -- in training capable Large Language Models (LLMs). To do so, we use the Task2Vec-based alignment coefficient, a quantitative measure of the similarity between two datasets, to quantify the impact of alignment between training data and evaluation data on downstream performance. In particular, we conduct controlled \textit{interventional} experiments for two settings: 1. the impact of increased alignment coefficients between various pre-training (pt) against evaluation datasets, and 2. the impact of increased alignment coefficients between domain specific fine-tuning (ft) against domain specific evaluation. The domain specific task we explore is Autoformalization -- the machine translation task between natural language and code for formal verification. In both settings, we find a strong, predictable negative correlation between the alignment coefficient of a model's training and evaluation data and the model's loss/perplexity on the respective downstream task. These findings suggest a re-evaluation of LLM training approaches, demonstrating the relevance of data alignment compared to data quantity, especially in specialized downstream tasks such as Autoformalization.
☆ The Theater Stage as Laboratory: Review of Real-Time Comedy LLM Systems for Live Performance COLING 2025
In this position paper, we review the eclectic recent history of academic and artistic works involving computational systems for humor generation, and focus specifically on live performance. We make the case that AI comedy should be evaluated in live conditions, in front of audiences sharing either physical or online spaces, and under real-time constraints. We further suggest that improvised comedy is therefore the perfect substrate for deploying and assessing computational humor systems. Using examples of successful AI-infused shows, we demonstrate that live performance raises three sets of challenges for computational humor generation: 1) questions around robotic embodiment, anthropomorphism and competition between humans and machines, 2) questions around comedic timing and the nature of audience interaction, and 3) questions about the human interpretation of seemingly absurd AI-generated humor. We argue that these questions impact the choice of methodologies for evaluating computational humor, as any such method needs to work around the constraints of live audiences and performance spaces. These interrogations also highlight different types of collaborative relationship of human comedians towards AI tools.
comment: 8 pages, 1st Workshop on Computational Humor (CHum), COLING 2025
☆ Selective Attention Merging for low resource tasks: A case study of Child ASR ICASSP 2025
While Speech Foundation Models (SFMs) excel in various speech tasks, their performance for low-resource tasks such as child Automatic Speech Recognition (ASR) is hampered by limited pretraining data. To address this, we explore different model merging techniques to leverage knowledge from models trained on larger, more diverse speech corpora. This paper also introduces Selective Attention (SA) Merge, a novel method that selectively merges task vectors from attention matrices to enhance SFM performance on low-resource tasks. Experiments on the MyST database show significant reductions in relative word error rate of up to 14%, outperforming existing model merging and data augmentation techniques. By combining data augmentation techniques with SA Merge, we achieve a new state-of-the-art WER of 8.69 on the MyST database for the Whisper-small model, highlighting the potential of SA Merge for improving low-resource ASR.
comment: To appear in ICASSP 2025
☆ Towards Zero-Shot & Explainable Video Description by Reasoning over Graphs of Events in Space and Time
In the current era of Machine Learning, Transformers have become the de facto approach across a variety of domains, such as computer vision and natural language processing. Transformer-based solutions are the backbone of current state-of-the-art methods for language generation, image and video classification, segmentation, action and object recognition, among many others. Interestingly enough, while these state-of-the-art methods produce impressive results in their respective domains, the problem of understanding the relationship between vision and language is still beyond our reach. In this work, we propose a common ground between vision and language based on events in space and time in an explainable and programmatic way, to connect learning-based vision and language state of the art models and provide a solution to the long standing problem of describing videos in natural language. We validate that our algorithmic approach is able to generate coherent, rich and relevant textual descriptions on videos collected from a variety of datasets, using both standard metrics (e.g. Bleu, ROUGE) and the modern LLM-as-a-Jury approach.
☆ Large Language Models For Text Classification: Case Study And Comprehensive Review
Unlocking the potential of Large Language Models (LLMs) in data classification represents a promising frontier in natural language processing. In this work, we evaluate the performance of different LLMs in comparison with state-of-the-art deep-learning and machine-learning models, in two different classification scenarios: i) the classification of employees' working locations based on job reviews posted online (multiclass classification), and 2) the classification of news articles as fake or not (binary classification). Our analysis encompasses a diverse range of language models differentiating in size, quantization, and architecture. We explore the impact of alternative prompting techniques and evaluate the models based on the weighted F1-score. Also, we examine the trade-off between performance (F1-score) and time (inference response time) for each language model to provide a more nuanced understanding of each model's practical applicability. Our work reveals significant variations in model responses based on the prompting strategies. We find that LLMs, particularly Llama3 and GPT-4, can outperform traditional methods in complex classification tasks, such as multiclass classification, though at the cost of longer inference times. In contrast, simpler ML models offer better performance-to-time trade-offs in simpler binary classification tasks.
☆ Tag&Tab: Pretraining Data Detection in Large Language Models Using Keyword-Based Membership Inference Attack
Large language models (LLMs) have become essential digital task assistance tools. Their training relies heavily on the collection of vast amounts of data, which may include copyright-protected or sensitive information. Recent studies on the detection of pretraining data in LLMs have primarily focused on sentence-level or paragraph-level membership inference attacks (MIAs), usually involving probability analysis of the target model prediction tokens. However, the proposed methods often demonstrate poor performance, specifically in terms of accuracy, failing to account for the semantic importance of textual content and word significance. To address these shortcomings, we propose Tag&Tab, a novel approach for detecting data that has been used as part of the LLM pretraining. Our method leverages advanced natural language processing (NLP) techniques to tag keywords in the input text - a process we term Tagging. Then, the LLM is used to obtain the probabilities of these keywords and calculate their average log-likelihood to determine input text membership, a process we refer to as Tabbing. Our experiments on three benchmark datasets (BookMIA, MIMIR, and the Pile) and several open-source LLMs of varying sizes demonstrate an average increase in the AUC scores ranging from 4.1% to 12.1% over state-of-the-art methods. Tag&Tab not only sets a new standard for data leakage detection in LLMs, but its outstanding performance is a testament to the importance of words in MIAs on LLMs.
☆ Jochre 3 and the Yiddish OCR corpus
We describe the construction of a publicly available Yiddish OCR Corpus, and describe and evaluate the open source OCR tool suite Jochre 3, including an Alto editor for corpus annotation, OCR software for Alto OCR layer generation, and a customizable OCR search engine. The current version of the Yiddish OCR corpus contains 658 pages, 186K tokens and 840K glyphs. The Jochre 3 OCR tool uses various fine-tuned YOLOv8 models for top-down page layout analysis, and a custom CNN network for glyph recognition. It attains a CER of 1.5% on our test corpus, far out-performing all other existing public models for Yiddish. We analyzed the full 660M word Yiddish Book Center with Jochre 3 OCR, and the new OCR is searchable through the Yiddish Book Center OCR search engine.
comment: 10 pages, 4 figures
☆ Religious Bias Landscape in Language and Text-to-Image Models: Analysis, Detection, and Debiasing Strategies
Note: This paper includes examples of potentially offensive content related to religious bias, presented solely for academic purposes. The widespread adoption of language models highlights the need for critical examinations of their inherent biases, particularly concerning religion. This study systematically investigates religious bias in both language models and text-to-image generation models, analyzing both open-source and closed-source systems. We construct approximately 400 unique, naturally occurring prompts to probe language models for religious bias across diverse tasks, including mask filling, prompt completion, and image generation. Our experiments reveal concerning instances of underlying stereotypes and biases associated disproportionately with certain religions. Additionally, we explore cross-domain biases, examining how religious bias intersects with demographic factors such as gender, age, and nationality. This study further evaluates the effectiveness of targeted debiasing techniques by employing corrective prompts designed to mitigate the identified biases. Our findings demonstrate that language models continue to exhibit significant biases in both text and image generation tasks, emphasizing the urgent need to develop fairer language models to achieve global acceptability.
☆ SEAL: Speaker Error Correction using Acoustic-conditioned Large Language Models ICASSP 2025
Speaker Diarization (SD) is a crucial component of modern end-to-end ASR pipelines. Traditional SD systems, which are typically audio-based and operate independently of ASR, often introduce speaker errors, particularly during speaker transitions and overlapping speech. Recently, language models including fine-tuned large language models (LLMs) have shown to be effective as a second-pass speaker error corrector by leveraging lexical context in the transcribed output. In this work, we introduce a novel acoustic conditioning approach to provide more fine-grained information from the acoustic diarizer to the LLM. We also show that a simpler constrained decoding strategy reduces LLM hallucinations, while avoiding complicated post-processing. Our approach significantly reduces the speaker error rates by 24-43% across Fisher, Callhome, and RT03-CTS datasets, compared to the first-pass Acoustic SD.
comment: Accepted at ICASSP 2025
☆ Ensemble of Large Language Models for Curated Labeling and Rating of Free-text Data
Free-text responses are commonly collected in psychological studies, providing rich qualitative insights that quantitative measures may not capture. Labeling curated topics of research interest in free-text data by multiple trained human coders is typically labor-intensive and time-consuming. Though large language models (LLMs) excel in language processing, LLM-assisted labeling techniques relying on closed-source LLMs cannot be directly applied to free-text data, without explicit consent for external use. In this study, we propose a framework of assembling locally-deployable LLMs to enhance the labeling of predetermined topics in free-text data under privacy constraints. Analogous to annotation by multiple human raters, this framework leverages the heterogeneity of diverse open-source LLMs. The ensemble approach seeks a balance between the agreement and disagreement across LLMs, guided by a relevancy scoring methodology that utilizes embedding distances between topic descriptions and LLMs' reasoning. We evaluated the ensemble approach using both publicly accessible Reddit data from eating disorder related forums, and free-text responses from eating disorder patients, both complemented by human annotations. We found that: (1) there is heterogeneity in the performance of labeling among same-sized LLMs, with some showing low sensitivity but high precision, while others exhibit high sensitivity but low precision. (2) Compared to individual LLMs, the ensemble of LLMs achieved the highest accuracy and optimal precision-sensitivity trade-off in predicting human annotations. (3) The relevancy scores across LLMs showed greater agreement than dichotomous labels, indicating that the relevancy scoring method effectively mitigates the heterogeneity in LLMs' labeling.
☆ OptiChat: Bridging Optimization Models and Practitioners with Large Language Models
Optimization models have been applied to solve a wide variety of decision-making problems. These models are usually developed by optimization experts but are used by practitioners without optimization expertise in various application domains. As a result, practitioners often struggle to interact with and draw useful conclusions from optimization models independently. To fill this gap, we introduce OptiChat, a natural language dialogue system designed to help practitioners interpret model formulation, diagnose infeasibility, analyze sensitivity, retrieve information, evaluate modifications, and provide counterfactual explanations. By augmenting large language models (LLMs) with functional calls and code generation tailored for optimization models, we enable seamless interaction and minimize the risk of hallucinations in OptiChat. We develop a new dataset to evaluate OptiChat's performance in explaining optimization models. Experiments demonstrate that OptiChat effectively bridges the gap between optimization models and practitioners, delivering autonomous, accurate, and instant responses.
☆ A Survey on Pedophile Attribution Techniques for Online Platforms
Reliance on anonymity in social media has increased its popularity on these platforms among all ages. The availability of public Wi-Fi networks has facilitated a vast variety of online content, including social media applications. Although anonymity and ease of access can be a convenient means of communication for their users, it is difficult to manage and protect its vulnerable users against sexual predators. Using an automated identification system that can attribute predators to their text would make the solution more attainable. In this survey, we provide a review of the methods of pedophile attribution used in social media platforms. We examine the effect of the size of the suspect set and the length of the text on the task of attribution. Moreover, we review the most-used datasets, features, classification techniques and performance measures for attributing sexual predators. We found that few studies have proposed tools to mitigate the risk of online sexual predators, but none of them can provide suspect attribution. Finally, we list several open research problems.
comment: 17 pages, 3 figures
☆ HALoGEN: Fantastic LLM Hallucinations and Where to Find Them
Despite their impressive ability to generate high-quality and fluent text, generative large language models (LLMs) also produce hallucinations: statements that are misaligned with established world knowledge or provided input context. However, measuring hallucination can be challenging, as having humans verify model generations on-the-fly is both expensive and time-consuming. In this work, we release HALoGEN, a comprehensive hallucination benchmark consisting of: (1) 10,923 prompts for generative models spanning nine domains including programming, scientific attribution, and summarization, and (2) automatic high-precision verifiers for each use case that decompose LLM generations into atomic units, and verify each unit against a high-quality knowledge source. We use this framework to evaluate ~150,000 generations from 14 language models, finding that even the best-performing models are riddled with hallucinations (sometimes up to 86% of generated atomic facts depending on the domain). We further define a novel error classification for LLM hallucinations based on whether they likely stem from incorrect recollection of training data (Type A errors), or incorrect knowledge in training data (Type B errors), or are fabrication (Type C errors). We hope our framework provides a foundation to enable the principled study of why generative models hallucinate, and advances the development of trustworthy large language models.
comment: Preprint
☆ Exploring Robustness of LLMs to Sociodemographically-Conditioned Paraphrasing
Large Language Models (LLMs) have shown impressive performance in various NLP tasks. However, there are concerns about their reliability in different domains of linguistic variations. Many works have proposed robustness evaluation measures for local adversarial attacks, but we need globally robust models unbiased to different language styles. We take a broader approach to explore a wider range of variations across sociodemographic dimensions to perform structured reliability tests on the reasoning capacity of language models. We extend the SocialIQA dataset to create diverse paraphrased sets conditioned on sociodemographic styles. The assessment aims to provide a deeper understanding of LLMs in (a) their capability of generating demographic paraphrases with engineered prompts and (b) their reasoning capabilities in real-world, complex language scenarios. We also explore measures such as perplexity, explainability, and ATOMIC performance of paraphrases for fine-grained reliability analysis of LLMs on these sets. We find that demographic-specific paraphrasing significantly impacts the performance of language models, indicating that the subtleties of language variations remain a significant challenge. The code and dataset will be made available for reproducibility and future research.
☆ Comparative Analysis of Efficient Adapter-Based Fine-Tuning of State-of-the-Art Transformer Models
In this work, we investigate the efficacy of various adapter architectures on supervised binary classification tasks from the SuperGLUE benchmark as well as a supervised multi-class news category classification task from Kaggle. Specifically, we compare classification performance and time complexity of three transformer models, namely DistilBERT, ELECTRA, and BART, using conventional fine-tuning as well as nine state-of-the-art (SoTA) adapter architectures. Our analysis reveals performance differences across adapter architectures, highlighting their ability to achieve comparable or better performance relative to fine-tuning at a fraction of the training time. Similar results are observed on the new classification task, further supporting our findings and demonstrating adapters as efficient and flexible alternatives to fine-tuning. This study provides valuable insights and guidelines for selecting and implementing adapters in diverse natural language processing (NLP) applications.
☆ Towards Best Practices for Open Datasets for LLM Training
Many AI companies are training their large language models (LLMs) on data without the permission of the copyright owners. The permissibility of doing so varies by jurisdiction: in countries like the EU and Japan, this is allowed under certain restrictions, while in the United States, the legal landscape is more ambiguous. Regardless of the legal status, concerns from creative producers have led to several high-profile copyright lawsuits, and the threat of litigation is commonly cited as a reason for the recent trend towards minimizing the information shared about training datasets by both corporate and public interest actors. This trend in limiting data information causes harm by hindering transparency, accountability, and innovation in the broader ecosystem by denying researchers, auditors, and impacted individuals access to the information needed to understand AI models. While this could be mitigated by training language models on open access and public domain data, at the time of writing, there are no such models (trained at a meaningful scale) due to the substantial technical and sociological challenges in assembling the necessary corpus. These challenges include incomplete and unreliable metadata, the cost and complexity of digitizing physical records, and the diverse set of legal and technical skills required to ensure relevance and responsibility in a quickly changing landscape. Building towards a future where AI systems can be trained on openly licensed data that is responsibly curated and governed requires collaboration across legal, technical, and policy domains, along with investments in metadata standards, digitization, and fostering a culture of openness.
☆ Eliciting In-context Retrieval and Reasoning for Long-context Large Language Models
Recent advancements in long-context language models (LCLMs) promise to transform Retrieval-Augmented Generation (RAG) by simplifying pipelines. With their expanded context windows, LCLMs can process entire knowledge bases and perform retrieval and reasoning directly -- a capability we define as In-Context Retrieval and Reasoning (ICR^2). However, existing benchmarks like LOFT often overestimate LCLM performance by providing overly simplified contexts. To address this, we introduce ICR^2, a benchmark that evaluates LCLMs in more realistic scenarios by including confounding passages retrieved with strong retrievers. We then propose three methods to enhance LCLM performance: (1) retrieve-then-generate fine-tuning, (2) retrieval-attention-probing, which uses attention heads to filter and de-noise long contexts during decoding, and (3) joint retrieval head training alongside the generation head. Our evaluation of five well-known LCLMs on LOFT and ICR^2 demonstrates significant gains with our best approach applied to Mistral-7B: +17 and +15 points by Exact Match on LOFT, and +13 and +2 points on ICR^2, compared to vanilla RAG and supervised fine-tuning, respectively. It even outperforms GPT-4-Turbo on most tasks despite being a much smaller model.
☆ ASTRID -- An Automated and Scalable TRIaD for the Evaluation of RAG-based Clinical Question Answering Systems
Large Language Models (LLMs) have shown impressive potential in clinical question answering (QA), with Retrieval Augmented Generation (RAG) emerging as a leading approach for ensuring the factual accuracy of model responses. However, current automated RAG metrics perform poorly in clinical and conversational use cases. Using clinical human evaluations of responses is expensive, unscalable, and not conducive to the continuous iterative development of RAG systems. To address these challenges, we introduce ASTRID - an Automated and Scalable TRIaD for evaluating clinical QA systems leveraging RAG - consisting of three metrics: Context Relevance (CR), Refusal Accuracy (RA), and Conversational Faithfulness (CF). Our novel evaluation metric, CF, is designed to better capture the faithfulness of a model's response to the knowledge base without penalising conversational elements. To validate our triad, we curate a dataset of over 200 real-world patient questions posed to an LLM-based QA agent during surgical follow-up for cataract surgery - the highest volume operation in the world - augmented with clinician-selected questions for emergency, clinical, and non-clinical out-of-domain scenarios. We demonstrate that CF can predict human ratings of faithfulness better than existing definitions for conversational use cases. Furthermore, we show that evaluation using our triad consisting of CF, RA, and CR exhibits alignment with clinician assessment for inappropriate, harmful, or unhelpful responses. Finally, using nine different LLMs, we demonstrate that the three metrics can closely agree with human evaluations, highlighting the potential of these metrics for use in LLM-driven automated evaluation pipelines. We also publish the prompts and datasets for these experiments, providing valuable resources for further research and development.
comment: 29 pages
☆ ArithmAttack: Evaluating Robustness of LLMs to Noisy Context in Math Problem Solving
While Large Language Models (LLMs) have shown impressive capabilities in math problem-solving tasks, their robustness to noisy inputs is not well-studied. In this work, we propose ArithmAttack to examine how robust the LLMs are when they encounter noisy prompts that contain extra noise in the form of punctuation marks. While being easy to implement, ArithmAttack does not cause any information loss since words are not added or deleted from the context. We evaluate the robustness of seven LLMs, including LLama3, Mistral, and Mathstral, on noisy GSM8K and MultiArith datasets. Our experiments suggest that all the studied models show vulnerability to such noise, with more noise leading to poorer performances.
☆ CWEval: Outcome-driven Evaluation on Functionality and Security of LLM Code Generation
Large Language Models (LLMs) have significantly aided developers by generating or assisting in code writing, enhancing productivity across various tasks. While identifying incorrect code is often straightforward, detecting vulnerabilities in functionally correct code is more challenging, especially for developers with limited security knowledge, which poses considerable security risks of using LLM-generated code and underscores the need for robust evaluation benchmarks that assess both functional correctness and security. Current benchmarks like CyberSecEval and SecurityEval attempt to solve it but are hindered by unclear and impractical specifications, failing to assess both functionality and security accurately. To tackle these deficiencies, we introduce CWEval, a novel outcome-driven evaluation framework designed to enhance the evaluation of secure code generation by LLMs. This framework not only assesses code functionality but also its security simultaneously with high-quality task specifications and outcome-driven test oracles which provides high accuracy. Coupled with CWEval-bench, a multilingual, security-critical coding benchmark, CWEval provides a rigorous empirical security evaluation on LLM-generated code, overcoming previous benchmarks' shortcomings. Through our evaluations, CWEval reveals a notable portion of functional but insecure code produced by LLMs, and shows a serious inaccuracy of previous evaluations, ultimately contributing significantly to the field of secure code generation. We open-source our artifact at: https://github.com/Co1lin/CWEval .
comment: to be published in LLM4Code 2025
☆ OpenCSG Chinese Corpus: A Series of High-quality Chinese Datasets for LLM Training
Large language models (LLMs) have demonstrated remarkable capabilities, but their success heavily relies on the quality of pretraining corpora. For Chinese LLMs, the scarcity of high-quality Chinese datasets presents a significant challenge, often limiting their performance. To address this issue, we propose the OpenCSG Chinese Corpus, a series of high-quality datasets specifically designed for LLM pretraining, post-training, and fine-tuning. This corpus includes Fineweb-edu-chinese, Fineweb-edu-chinese-v2, Cosmopedia-chinese, and Smoltalk-chinese, each with distinct characteristics: Fineweb-edu datasets focus on filtered, high-quality content derived from diverse Chinese web sources; Cosmopedia-chinese provides synthetic, textbook-style data for knowledge-intensive training; and Smoltalk-chinese emphasizes stylistic and diverse chat-format data. The OpenCSG Chinese Corpus is characterized by its high-quality text, diverse coverage across domains, and scalable, reproducible data curation processes. Additionally, we conducted extensive experimental analyses, including evaluations on smaller parameter models, which demonstrated significant performance improvements in tasks such as C-Eval, showcasing the effectiveness of the corpus for training Chinese LLMs.
comment: The datasets are available on https://huggingface.co/collections/opencsg/chinese-fineweb-66cfed105f502ece8f29643e ; The code is on https://github.com/yuyijiong/fineweb-edu-chinese
☆ Potential and Perils of Large Language Models as Judges of Unstructured Textual Data
Rapid advancements in large language models have unlocked remarkable capabilities when it comes to processing and summarizing unstructured text data. This has implications for the analysis of rich, open-ended datasets, such as survey responses, where LLMs hold the promise of efficiently distilling key themes and sentiments. However, as organizations increasingly turn to these powerful AI systems to make sense of textual feedback, a critical question arises, can we trust LLMs to accurately represent the perspectives contained within these text based datasets? While LLMs excel at generating human-like summaries, there is a risk that their outputs may inadvertently diverge from the true substance of the original responses. Discrepancies between the LLM-generated outputs and the actual themes present in the data could lead to flawed decision-making, with far-reaching consequences for organizations. This research investigates the effectiveness of LLMs as judge models to evaluate the thematic alignment of summaries generated by other LLMs. We utilized an Anthropic Claude model to generate thematic summaries from open-ended survey responses, with Amazon's Titan Express, Nova Pro, and Meta's Llama serving as LLM judges. The LLM-as-judge approach was compared to human evaluations using Cohen's kappa, Spearman's rho, and Krippendorff's alpha, validating a scalable alternative to traditional human centric evaluation methods. Our findings reveal that while LLMs as judges offer a scalable solution comparable to human raters, humans may still excel at detecting subtle, context-specific nuances. This research contributes to the growing body of knowledge on AI assisted text analysis. We discuss limitations and provide recommendations for future research, emphasizing the need for careful consideration when generalizing LLM judge models across various contexts and use cases.
comment: 11 pages, 1 appendix
☆ Refusal Behavior in Large Language Models: A Nonlinear Perspective
Refusal behavior in large language models (LLMs) enables them to decline responding to harmful, unethical, or inappropriate prompts, ensuring alignment with ethical standards. This paper investigates refusal behavior across six LLMs from three architectural families. We challenge the assumption of refusal as a linear phenomenon by employing dimensionality reduction techniques, including PCA, t-SNE, and UMAP. Our results reveal that refusal mechanisms exhibit nonlinear, multidimensional characteristics that vary by model architecture and layer. These findings highlight the need for nonlinear interpretability to improve alignment research and inform safer AI deployment strategies.
☆ In-situ graph reasoning and knowledge expansion using Graph-PReFLexOR
The pursuit of automated scientific discovery has fueled progress from symbolic logic to modern AI, forging new frontiers in reasoning and pattern recognition. Transformers function as potential systems, where every possible relationship remains latent potentiality until tasks impose constraints, akin to measurement. Yet, refining their sampling requires more than probabilistic selection: solutions must conform to specific structures or rules, ensuring consistency and the invocation of general principles. We present Graph-PReFLexOR (Graph-based Preference-based Recursive Language Modeling for Exploratory Optimization of Reasoning), a framework that combines graph reasoning with symbolic abstraction to dynamically expand domain knowledge. Inspired by reinforcement learning, Graph-PReFLexOR defines reasoning as a structured mapping, where tasks yield knowledge graphs, abstract patterns, and ultimately, final answers. Inspired by category theory, it encodes concepts as nodes and their relationships as edges, supporting hierarchical inference and adaptive learning through isomorphic representations. Demonstrations include hypothesis generation, materials design, and creative reasoning, such as discovering relationships between mythological concepts like 'thin places' with materials science. We propose a 'knowledge garden growth' strategy that integrates insights across domains, promoting interdisciplinary connections. Results with a 3-billion-parameter Graph-PReFLexOR model show superior reasoning depth and adaptability, underscoring the potential for transparent, multidisciplinary AI-driven discovery. It lays the groundwork for general autonomous reasoning solutions.
☆ Dynamic Multimodal Sentiment Analysis: Leveraging Cross-Modal Attention for Enabled Classification
This paper explores the development of a multimodal sentiment analysis model that integrates text, audio, and visual data to enhance sentiment classification. The goal is to improve emotion detection by capturing the complex interactions between these modalities, thereby enabling more accurate and nuanced sentiment interpretation. The study evaluates three feature fusion strategies -- late stage fusion, early stage fusion, and multi-headed attention -- within a transformer-based architecture. Experiments were conducted using the CMU-MOSEI dataset, which includes synchronized text, audio, and visual inputs labeled with sentiment scores. Results show that early stage fusion significantly outperforms late stage fusion, achieving an accuracy of 71.87\%, while the multi-headed attention approach offers marginal improvement, reaching 72.39\%. The findings suggest that integrating modalities early in the process enhances sentiment classification, while attention mechanisms may have limited impact within the current framework. Future work will focus on refining feature fusion techniques, incorporating temporal data, and exploring dynamic feature weighting to further improve model performance.
☆ Optimizing Speech Multi-View Feature Fusion through Conditional Computation ICASSP 2025
Recent advancements have highlighted the efficacy of self-supervised learning (SSL) features in various speech-related tasks, providing lightweight and versatile multi-view speech representations. However, our study reveals that while SSL features expedite model convergence, they conflict with traditional spectral features like FBanks in terms of update directions. In response, we propose a novel generalized feature fusion framework grounded in conditional computation, featuring a gradient-sensitive gating network and a multi-stage dropout strategy. This framework mitigates feature conflicts and bolsters model robustness to multi-view input features. By integrating SSL and spectral features, our approach accelerates convergence and maintains performance on par with spectral models across multiple speech translation tasks on the MUSTC dataset.
comment: ICASSP 2025
☆ Exploring Narrative Clustering in Large Language Models: A Layerwise Analysis of BERT
This study investigates the internal mechanisms of BERT, a transformer-based large language model, with a focus on its ability to cluster narrative content and authorial style across its layers. Using a dataset of narratives developed via GPT-4, featuring diverse semantic content and stylistic variations, we analyze BERT's layerwise activations to uncover patterns of localized neural processing. Through dimensionality reduction techniques such as Principal Component Analysis (PCA) and Multidimensional Scaling (MDS), we reveal that BERT exhibits strong clustering based on narrative content in its later layers, with progressively compact and distinct clusters. While strong stylistic clustering might occur when narratives are rephrased into different text types (e.g., fables, sci-fi, kids' stories), minimal clustering is observed for authorial style specific to individual writers. These findings highlight BERT's prioritization of semantic content over stylistic features, offering insights into its representational capabilities and processing hierarchy. This study contributes to understanding how transformer models like BERT encode linguistic information, paving the way for future interdisciplinary research in artificial intelligence and cognitive neuroscience.
comment: arXiv admin note: text overlap with arXiv:2408.03062, arXiv:2408.04270, arXiv:2307.01577
☆ READ: Reinforcement-based Adversarial Learning for Text Classification with Limited Labeled Data
Pre-trained transformer models such as BERT have shown massive gains across many text classification tasks. However, these models usually need enormous labeled data to achieve impressive performances. Obtaining labeled data is often expensive and time-consuming, whereas collecting unlabeled data using some heuristics is relatively much cheaper for any task. Therefore, this paper proposes a method that encapsulates reinforcement learning-based text generation and semi-supervised adversarial learning approaches in a novel way to improve the model's performance. Our method READ, Reinforcement-based Adversarial learning, utilizes an unlabeled dataset to generate diverse synthetic text through reinforcement learning, improving the model's generalization capability using adversarial learning. Our experimental results show that READ outperforms the existing state-of-art methods on multiple datasets.
☆ TriAdaptLoRA: Brain-Inspired Triangular Adaptive Low-Rank Adaptation for Parameter-Efficient Fine-Tuning
The fine-tuning of Large Language Models (LLMs) is pivotal for achieving optimal performance across diverse downstream tasks. However, while full fine-tuning delivers superior results, it entails significant computational and resource costs. Parameter-Efficient Fine-Tuning (PEFT) methods, such as LoRA, address these challenges by reducing the number of trainable parameters, but they often struggle with rank adjustment efficiency and task-specific adaptability. We propose Triangular Adaptive Low-Rank Adaptation (TriAdaptLoRA), a novel PEFT framework inspired by neuroscience principles, which dynamically optimizes the allocation of trainable parameters. TriAdaptLoRA introduces three key innovations: 1) a triangular split of transformation matrices into lower and upper triangular components to maximize parameter utilization, 2) a parameter importance metric based on normalized Frobenius norms for efficient adaptation, and 3) an adaptive rank-growth strategy governed by dynamic thresholds, allowing flexible parameter allocation across training steps. Experiments conducted on a variety of natural language understanding and generation tasks demonstrate that TriAdaptLoRA consistently outperforms existing PEFT methods. It achieves superior performance, enhanced stability, and reduced computational overhead, particularly under linear threshold-driven rank growth. These results highlight its efficacy as a scalable and resource-efficient solution for fine-tuning LLMs.
☆ Formalising lexical and syntactic diversity for data sampling in French
Diversity is an important property of datasets and sampling data for diversity is useful in dataset creation. Finding the optimally diverse sample is expensive, we therefore present a heuristic significantly increasing diversity relative to random sampling. We also explore whether different kinds of diversity -- lexical and syntactic -- correlate, with the purpose of sampling for expensive syntactic diversity through inexpensive lexical diversity. We find that correlations fluctuate with different datasets and versions of diversity measures. This shows that an arbitrarily chosen measure may fall short of capturing diversity-related properties of datasets.
☆ "Wait, did you mean the doctor?": Collecting a Dialogue Corpus for Topical Analysis
Dialogue is at the core of human behaviour and being able to identify the topic at hand is crucial to take part in conversation. Yet, there are few accounts of the topical organisation in casual dialogue and of how people recognise the current topic in the literature. Moreover, analysing topics in dialogue requires conversations long enough to contain several topics and types of topic shifts. Such data is complicated to collect and annotate. In this paper we present a dialogue collection experiment which aims to build a corpus suitable for topical analysis. We will carry out the collection with a messaging tool we developed.
☆ Gandalf the Red: Adaptive Security for LLMs
Current evaluations of defenses against prompt attacks in large language model (LLM) applications often overlook two critical factors: the dynamic nature of adversarial behavior and the usability penalties imposed on legitimate users by restrictive defenses. We propose D-SEC (Dynamic Security Utility Threat Model), which explicitly separates attackers from legitimate users, models multi-step interactions, and rigorously expresses the security-utility in an optimizable form. We further address the shortcomings in existing evaluations by introducing Gandalf, a crowd-sourced, gamified red-teaming platform designed to generate realistic, adaptive attack datasets. Using Gandalf, we collect and release a dataset of 279k prompt attacks. Complemented by benign user data, our analysis reveals the interplay between security and utility, showing that defenses integrated in the LLM (e.g., system prompts) can degrade usability even without blocking requests. We demonstrate that restricted application domains, defense-in-depth, and adaptive defenses are effective strategies for building secure and useful LLM applications. Code is available at \href{https://github.com/lakeraai/dsec-gandalf}{\texttt{https://github.com/lakeraai/dsec-gandalf}}.
comment: Niklas Pfister, V\'aclav Volhejn and Manuel Knott contributed equally
☆ Exploring Aviation Incident Narratives Using Topic Modeling and Clustering Techniques
Aviation safety is a global concern, requiring detailed investigations into incidents to understand contributing factors comprehensively. This study uses the National Transportation Safety Board (NTSB) dataset. It applies advanced natural language processing (NLP) techniques, including Latent Dirichlet Allocation (LDA), Non-Negative Matrix Factorization (NMF), Latent Semantic Analysis (LSA), Probabilistic Latent Semantic Analysis (pLSA), and K-means clustering. The main objectives are identifying latent themes, exploring semantic relationships, assessing probabilistic connections, and cluster incidents based on shared characteristics. This research contributes to aviation safety by providing insights into incident narratives and demonstrating the versatility of NLP and topic modelling techniques in extracting valuable information from complex datasets. The results, including topics identified from various techniques, provide an understanding of recurring themes. Comparative analysis reveals that LDA performed best with a coherence value of 0.597, pLSA of 0.583, LSA of 0.542, and NMF of 0.437. K-means clustering further reveals commonalities and unique insights into incident narratives. In conclusion, this study uncovers latent patterns and thematic structures within incident narratives, offering a comparative analysis of multiple-topic modelling techniques. Future research avenues include exploring temporal patterns, incorporating additional datasets, and developing predictive models for early identification of safety issues. This research lays the groundwork for enhancing the understanding and improvement of aviation safety by utilising the wealth of information embedded in incident narratives.
☆ Aviation Safety Enhancement via NLP & Deep Learning: Classifying Flight Phases in ATSB Safety Reports
Aviation safety is paramount, demanding precise analysis of safety occurrences during different flight phases. This study employs Natural Language Processing (NLP) and Deep Learning models, including LSTM, CNN, Bidirectional LSTM (BLSTM), and simple Recurrent Neural Networks (sRNN), to classify flight phases in safety reports from the Australian Transport Safety Bureau (ATSB). The models exhibited high accuracy, precision, recall, and F1 scores, with LSTM achieving the highest performance of 87%, 88%, 87%, and 88%, respectively. This performance highlights their effectiveness in automating safety occurrence analysis. The integration of NLP and Deep Learning technologies promises transformative enhancements in aviation safety analysis, enabling targeted safety measures and streamlined report handling.
comment: NLP, Aviation Safety, ATSB, Deep learning, Flight phase. arXiv admin note: substantial text overlap with arXiv:2501.01694
☆ GRAPHMOE: Amplifying Cognitive Depth of Mixture-of-Experts Network via Introducing Self-Rethinking Mechanism
Traditional Mixture-of-Experts (MoE) networks benefit from utilizing multiple smaller expert models as opposed to a single large network. However, these experts typically operate independently, leaving a question open about whether interconnecting these models could enhance the performance of MoE networks. In response, we introduce GRAPHMOE, a novel method aimed at augmenting the cognitive depth of language models via a self-rethinking mechanism constructed on Pseudo GraphMoE networks. GRAPHMOE employs a recurrent routing strategy to simulate iterative thinking steps, thereby facilitating the flow of information among expert nodes. We implement the GRAPHMOE architecture using Low-Rank Adaptation techniques (LoRA) and conduct extensive experiments on various benchmark datasets. The experimental results reveal that GRAPHMOE outperforms other LoRA based models, achieving state-of-the-art (SOTA) performance. Additionally, this study explores a novel recurrent routing strategy that may inspire further advancements in enhancing the reasoning capabilities of language models.
comment: 10 pages
☆ Iterative Label Refinement Matters More than Preference Optimization under Weak Supervision
Language model (LM) post-training relies on two stages of human supervision: task demonstrations for supervised finetuning (SFT), followed by preference comparisons for reinforcement learning from human feedback (RLHF). As LMs become more capable, the tasks they are given become harder to supervise. Will post-training remain effective under unreliable supervision? To test this, we simulate unreliable demonstrations and comparison feedback using small LMs and time-constrained humans. We find that in the presence of unreliable supervision, SFT still retains some effectiveness, but DPO (a common RLHF algorithm) fails to improve the model beyond SFT. To address this, we propose iterative label refinement (ILR) as an alternative to RLHF. ILR improves the SFT data by using comparison feedback to decide whether human demonstrations should be replaced by model-generated alternatives, then retrains the model via SFT on the updated data. SFT+ILR outperforms SFT+DPO on several tasks with unreliable supervision (math, coding, and safe instruction-following). Our findings suggest that as LMs are used for complex tasks where human supervision is unreliable, RLHF may no longer be the best use of human comparison feedback; instead, it is better to direct feedback towards improving the training data rather than continually training the model. Our code and data are available at https://github.com/helloelwin/iterative-label-refinement.
comment: 22 pages, 10 figures
☆ Continual Learning with Embedding Layer Surgery and Task-wise Beam Search using Whisper
Current Multilingual ASR models only support a fraction of the world's languages. Continual Learning (CL) aims to tackle this problem by adding new languages to pre-trained models while avoiding the loss of performance on existing languages, also known as Catastrophic Forgetting (CF). However, existing CL methods overlook the adaptation of the token embedding lookup table at the decoder, despite its significant contribution to CF. We propose Embedding Layer Surgery where separate copies of the token embeddings are created for each new languages, and one of the copies is selected to replace the old languages embeddings when transcribing the corresponding new language. Unfortunately, this approach means LID errors also cause incorrect ASR embedding selection. Our Task-wise Beam Search allows self-correction for such mistakes. By adapting Whisper to 10 hours of data for each of 10 unseen languages from Common Voice, results show that our method reduces the Average WER (AWER) of pre-trained languages from 14.2% to 11.9% compared with Experience Replay, without compromising the AWER of the unseen languages.
comment: Published in 2024 IEEE Spoken Language Technology Workshop
☆ ReARTeR: Retrieval-Augmented Reasoning with Trustworthy Process Rewarding
Retrieval-Augmented Generation (RAG) systems for Large Language Models (LLMs) hold promise in knowledge-intensive tasks but face limitations in complex multi-step reasoning. While recent methods have integrated RAG with chain-of-thought reasoning or test-time search using Process Reward Models (PRMs), these approaches encounter challenges such as a lack of explanations, bias in PRM training data, early-step bias in PRM scores, and insufficient post-training optimization of reasoning potential. To address these issues, we propose Retrieval-Augmented Reasoning through Trustworthy Process Rewarding (ReARTeR), a framework that enhances RAG systems' reasoning capabilities through post-training and test-time scaling. At test time, ReARTeR introduces Trustworthy Process Rewarding via a Process Reward Model for accurate scalar scoring and a Process Explanation Model (PEM) for generating natural language explanations, enabling step refinement. During post-training, it utilizes Monte Carlo Tree Search guided by Trustworthy Process Rewarding to collect high-quality step-level preference data, optimized through Iterative Preference Optimization. ReARTeR addresses three core challenges: (1) misalignment between PRM and PEM, tackled through off-policy preference learning; (2) bias in PRM training data, mitigated by balanced annotation methods and stronger annotations for challenging examples; and (3) early-step bias in PRM, resolved through a temporal-difference-based look-ahead search strategy. Experimental results on multi-step reasoning benchmarks demonstrate significant improvements, underscoring ReARTeR's potential to advance the reasoning capabilities of RAG systems.
comment: 11 pages, 5 figures
☆ Optimizing Language Models for Grammatical Acceptability: A Comparative Study of Fine-Tuning Techniques
This study explores the fine-tuning (FT) of the Open Pre-trained Transformer (OPT-125M) for grammatical acceptability tasks using the CoLA dataset. By comparing Vanilla-Fine-Tuning (VFT), Pattern-Based-Fine-Tuning (PBFT), and Parameter-Efficient Fine-Tuning techniques (PEFT) like Low-Rank Adaptation (LoRA), we demonstrate significant improvements in computational efficiency while maintaining high accuracy. Our experiments reveal that while VFT achieves the highest accuracy (81.2%), LoRA enhancing FT by reducing memory usage and iteration time by more than 50%, and increases accuracy in PBFT case. Context Distillation (CD), though computationally efficient, underperformed with accuracy around 31%. Our findings contribute to democratizing access to large language models (LLM) by reducing computational barriers.
☆ Reasoning with Graphs: Structuring Implicit Knowledge to Enhance LLMs Reasoning
Large language models (LLMs) have demonstrated remarkable success across a wide range of tasks; however, they still encounter challenges in reasoning tasks that require understanding and inferring relationships between distinct pieces of information within text sequences. This challenge is particularly pronounced in tasks involving multi-step processes, such as logical reasoning and multi-hop question answering, where understanding implicit relationships between entities and leveraging multi-hop connections in the given context are crucial. Graphs, as fundamental data structures, explicitly represent pairwise relationships between entities, thereby offering the potential to enhance LLMs' reasoning capabilities. External graphs have proven effective in supporting LLMs across multiple tasks. However, in many reasoning tasks, no pre-existing graph structure is provided. Can we structure implicit knowledge derived from context into graphs to assist LLMs in reasoning? In this paper, we propose Reasoning with Graphs (RwG) by first constructing explicit graphs from the context and then leveraging these graphs to enhance LLM reasoning performance on reasoning tasks. Extensive experiments demonstrate the effectiveness of the proposed method in improving both logical reasoning and multi-hop question answering tasks.
☆ Real-time Verification and Refinement of Language Model Text Generation
Large language models (LLMs) have shown remarkable performance across a wide range of natural language tasks. However, a critical challenge remains in that they sometimes generate factually incorrect answers. To address this, while many previous work has focused on identifying errors in their generation and further refining them, they are slow in deployment since they are designed to verify the response from LLMs only after their entire generation (from the first to last tokens) is done. Further, we observe that once LLMs generate incorrect tokens early on, there is a higher likelihood that subsequent tokens will also be factually incorrect. To this end, in this work, we propose Streaming-VR (Streaming Verification and Refinement), a novel approach designed to enhance the efficiency of verification and refinement of LLM outputs. Specifically, the proposed Streaming-VR enables on-the-fly verification and correction of tokens as they are being generated, similar to a streaming process, ensuring that each subset of tokens is checked and refined in real-time by another LLM as the LLM constructs its response. Through comprehensive evaluations on multiple datasets, we demonstrate that our approach not only enhances the factual accuracy of LLMs, but also offers a more efficient solution compared to prior refinement methods.
comment: Preprint
☆ A Multi-Encoder Frozen-Decoder Approach for Fine-Tuning Large Language Models
Among parameter-efficient fine-tuning methods, freezing has emerged as a popular strategy for speeding up training, reducing catastrophic forgetting, and improving downstream performance. We investigate the impact of freezing the decoder in a multi-task setup comprising diverse natural language tasks, aiming to reduce deployment overhead and enhance portability to novel tasks. Our experiments, conducted by fine-tuning both individual and multi-task setups on the AlexaTM model, reveal that freezing decoders is highly effective for tasks with natural language outputs and mitigates catastrophic forgetting in multilingual tasks. However, we find that pairing frozen decoders with a larger model can effectively maintain or even enhance performance in structured and QA tasks, making it a viable strategy for a broader range of task types.
☆ Agent-Centric Projection of Prompting Techniques and Implications for Synthetic Training Data for Large Language Models
Recent advances in prompting techniques and multi-agent systems for Large Language Models (LLMs) have produced increasingly complex approaches. However, we lack a framework for characterizing and comparing prompting techniques or understanding their relationship to multi-agent LLM systems. This position paper introduces and explains the concepts of linear contexts (a single, continuous sequence of interactions) and non-linear contexts (branching or multi-path) in LLM systems. These concepts enable the development of an agent-centric projection of prompting techniques, a framework that can reveal deep connections between prompting strategies and multi-agent systems. We propose three conjectures based on this framework: (1) results from non-linear prompting techniques can predict outcomes in equivalent multi-agent systems, (2) multi-agent system architectures can be replicated through single-LLM prompting techniques that simulate equivalent interaction patterns, and (3) these equivalences suggest novel approaches for generating synthetic training data. We argue that this perspective enables systematic cross-pollination of research findings between prompting and multi-agent domains, while providing new directions for improving both the design and training of future LLM systems.
comment: 8 pages, 5 figures. Accepted at ICAART 2025. Derived from an early draft at 2312.17601. arXiv admin note: substantial text overlap with arXiv:2312.17601
☆ Talk to Right Specialists: Routing and Planning in Multi-agent System for Question Answering
Leveraging large language models (LLMs), an agent can utilize retrieval-augmented generation (RAG) techniques to integrate external knowledge and increase the reliability of its responses. Current RAG-based agents integrate single, domain-specific knowledge sources, limiting their ability and leading to hallucinated or inaccurate responses when addressing cross-domain queries. Integrating multiple knowledge bases into a unified RAG-based agent raises significant challenges, including increased retrieval overhead and data sovereignty when sensitive data is involved. In this work, we propose RopMura, a novel multi-agent system that addresses these limitations by incorporating highly efficient routing and planning mechanisms. RopMura features two key components: a router that intelligently selects the most relevant agents based on knowledge boundaries and a planner that decomposes complex multi-hop queries into manageable steps, allowing for coordinating cross-domain responses. Experimental results demonstrate that RopMura effectively handles both single-hop and multi-hop queries, with the routing mechanism enabling precise answers for single-hop queries and the combined routing and planning mechanisms achieving accurate, multi-step resolutions for complex queries.
comment: Work In Progress
☆ Parameter-Inverted Image Pyramid Networks for Visual Perception and Multimodal Understanding
Image pyramids are widely adopted in top-performing methods to obtain multi-scale features for precise visual perception and understanding. However, current image pyramids use the same large-scale model to process multiple resolutions of images, leading to significant computational cost. To address this challenge, we propose a novel network architecture, called Parameter-Inverted Image Pyramid Networks (PIIP). Specifically, PIIP uses pretrained models (ViTs or CNNs) as branches to process multi-scale images, where images of higher resolutions are processed by smaller network branches to balance computational cost and performance. To integrate information from different spatial scales, we further propose a novel cross-branch feature interaction mechanism. To validate PIIP, we apply it to various perception models and a representative multimodal large language model called LLaVA, and conduct extensive experiments on various tasks such as object detection, segmentation, image classification and multimodal understanding. PIIP achieves superior performance compared to single-branch and existing multi-resolution approaches with lower computational cost. When applied to InternViT-6B, a large-scale vision foundation model, PIIP can improve its performance by 1%-2% on detection and segmentation with only 40%-60% of the original computation, finally achieving 60.0 box AP on MS COCO and 59.7 mIoU on ADE20K. For multimodal understanding, our PIIP-LLaVA achieves 73.0% accuracy on TextVQA and 74.5% on MMBench with only 2.8M training data. Our code is released at https://github.com/OpenGVLab/PIIP.
☆ Large Language Models for Knowledge Graph Embedding Techniques, Methods, and Challenges: A Survey
Large Language Models (LLMs) have attracted a lot of attention in various fields due to their superior performance, aiming to train hundreds of millions or more parameters on large amounts of text data to understand and generate natural language. As the superior performance of LLMs becomes apparent, they are increasingly being applied to knowledge graph embedding (KGE) related tasks to improve the processing results. As a deep learning model in the field of Natural Language Processing (NLP), it learns a large amount of textual data to predict the next word or generate content related to a given text. However, LLMs have recently been invoked to varying degrees in different types of KGE related scenarios such as multi-modal KGE and open KGE according to their task characteristics. In this paper, we investigate a wide range of approaches for performing LLMs-related tasks in different types of KGE scenarios. To better compare the various approaches, we summarize each KGE scenario in a classification. In addition to the categorization methods, we provide a tabular overview of the methods and their source code links for a more direct comparison. In the article we also discuss the applications in which the methods are mainly used and suggest several forward-looking directions for the development of this new research area.
♻ ☆ Multigenre AI-powered Story Composition
This paper shows how to construct genre patterns, whose purpose is to guide interactive story composition in a way that enforces thematic consistency. To start the discussion we argue, based on previous seminal works, for the existence of five fundamental genres, namely comedy, romance - in the sense of epic plots, flourishing since the twelfth century -, tragedy, satire, and mystery. To construct the patterns, a simple two-phase process is employed: first retrieving examples that match our genre characterizations, and then applying a form of most specific generalization to the groups of examples in order to find their commonalities. In both phases, AI agents are instrumental, with our PatternTeller prototype being called to operate the story composition process, offering the opportunity to generate stories from a given premise of the user, to be developed under the guidance of the chosen pattern and trying to accommodate the user's suggestions along the composition stages.
comment: Added publication details to references that were published after the submission of the previous version (references [18] and [19])
♻ ☆ Empowering Persian LLMs for Instruction Following: A Novel Dataset and Training Approach
Instruction-tuned large language models have demonstrated remarkable capabilities in following human instructions across various domains. However, their proficiency remains notably deficient in many low-resource languages. To address this challenge, we begin by introducing FarsInstruct a comprehensive instruction dataset designed to enhance the instruction following ability of large language models specifically for the Persian language a significant yet underrepresented language globally. FarsInstruct encompasses a wide range of task types and datasets, each containing a mix of straightforward to complex manual written instructions, as well as translations from the Public Pool of Prompts, ensuring a rich linguistic and cultural representation. Furthermore, we introduce Co-CoLA, a framework designed to enhance the multi-task adaptability of LoRA-tuned models. Through extensive experimental analyses, our study showcases the effectiveness of the FarsInstruct dataset coupled with training by the Co-CoLA framework, in improving the performance of large language models within the Persian context. As of the current writing, FarsInstruct comprises 197 templates across 21 distinct datasets, and we intend to update it consistently, thus augmenting its applicability.
♻ ☆ U-MATH: A University-Level Benchmark for Evaluating Mathematical Skills in LLMs
The current evaluation of mathematical skills in LLMs is limited, as existing benchmarks are either relatively small, primarily focus on elementary and high-school problems, or lack diversity in topics. Additionally, the inclusion of visual elements in tasks remains largely under-explored. To address these gaps, we introduce U-MATH, a novel benchmark of 1,100 unpublished open-ended university-level problems sourced from teaching materials. It is balanced across six core subjects, with 20% of multimodal problems. Given the open-ended nature of U-MATH problems, we employ an LLM to judge the correctness of generated solutions. To this end, we release $\mu$-MATH, a dataset to evaluate the LLMs' capabilities in judging solutions. The evaluation of general domain, math-specific, and multimodal LLMs highlights the challenges presented by U-MATH. Our findings reveal that LLMs achieve a maximum accuracy of only 63% on text-based tasks, with even lower 45% on visual problems. The solution assessment proves challenging for LLMs, with the best LLM judge having an F1-score of 80% on $\mu$-MATH.
♻ ☆ A Multi-way Parallel Named Entity Annotated Corpus for English, Tamil and Sinhala
This paper presents a multi-way parallel English-Tamil-Sinhala corpus annotated with Named Entities (NEs), where Sinhala and Tamil are low-resource languages. Using pre-trained multilingual Language Models (mLMs), we establish new benchmark Named Entity Recognition (NER) results on this dataset for Sinhala and Tamil. We also carry out a detailed investigation on the NER capabilities of different types of mLMs. Finally, we demonstrate the utility of our NER system on a low-resource Neural Machine Translation (NMT) task. Our dataset is publicly released: https://github.com/suralk/multiNER.
♻ ☆ CriSPO: Multi-Aspect Critique-Suggestion-guided Automatic Prompt Optimization for Text Generation AAAI-2025
Existing automatic prompt engineering methods are typically designed for discriminative tasks, where new task prompts are iteratively refined with limited feedback from a single metric reflecting a single aspect. However, these approaches are suboptimal for generative tasks, which require more nuanced guidance beyond a single numeric metric to improve the prompt and optimize multiple aspects of the generated text. To address these challenges, we propose a novel multi-aspect Critique-Suggestion-guided automatic Prompt Optimization (CriSPO) approach. CriSPO introduces a critique-suggestion module as its core component. This module spontaneously discovers aspects, and compares generated and reference texts across these aspects, providing specific suggestions for prompt modification. These clear critiques and actionable suggestions guide a receptive optimizer module to make more substantial changes, exploring a broader and more effective search space. To further improve CriSPO with multi-metric optimization, we introduce an Automatic Suffix Tuning (AST) extension to enhance the performance of task prompts across multiple metrics. We evaluate CriSPO on 4 state-of-the-art LLMs across 4 summarization and 5 QA datasets. Extensive experiments show 3-4% ROUGE score improvement on summarization and substantial improvement of various metrics on QA. Code available at https://github.com/amazon-science/crispo
comment: Accepted to AAAI-2025
♻ ☆ Bilingual Evaluation of Language Models on General Knowledge in University Entrance Exams with Minimal Contamination
In this article we present UNED-ACCESS 2024, a bilingual dataset that consists of 1003 multiple-choice questions of university entrance level exams in Spanish and English. Questions are originally formulated in Spanish and translated manually into English, and have not ever been publicly released. A selection of current open-source and proprietary models are evaluated in a uniform zero-shot experimental setting both on the UNED-ACCESS 2024 dataset and on an equivalent subset of MMLU questions. Results show that (i) reasoning questions are challenging for models, (ii) smaller models perform worse than larger models and degrade faster in Spanish than in English and (iii) the performance gap between languages is negligible for the best models and grows up to 37% for smaller models. Model ranking on UNED-ACCESS 2024 is almost identical in English and Spanish, and has also a high correlation (0.98 Pearson) with ranking on MMLU, suggesting that a small dataset is sufficiently diverse and representative to measure performance by discipline.
♻ ☆ HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models NeurIPS 2024
In order to thrive in hostile and ever-changing natural environments, mammalian brains evolved to store large amounts of knowledge about the world and continually integrate new information while avoiding catastrophic forgetting. Despite the impressive accomplishments, large language models (LLMs), even with retrieval-augmented generation (RAG), still struggle to efficiently and effectively integrate a large amount of new experiences after pre-training. In this work, we introduce HippoRAG, a novel retrieval framework inspired by the hippocampal indexing theory of human long-term memory to enable deeper and more efficient knowledge integration over new experiences. HippoRAG synergistically orchestrates LLMs, knowledge graphs, and the Personalized PageRank algorithm to mimic the different roles of neocortex and hippocampus in human memory. We compare HippoRAG with existing RAG methods on multi-hop question answering and show that our method outperforms the state-of-the-art methods remarkably, by up to 20%. Single-step retrieval with HippoRAG achieves comparable or better performance than iterative retrieval like IRCoT while being 10-30 times cheaper and 6-13 times faster, and integrating HippoRAG into IRCoT brings further substantial gains. Finally, we show that our method can tackle new types of scenarios that are out of reach of existing methods. Code and data are available at https://github.com/OSU-NLP-Group/HippoRAG.
comment: NeurIPS 2024. Code and data: https://github.com/OSU-NLP-Group/HippoRAG
♻ ☆ Text-guided Image Restoration and Semantic Enhancement for Text-to-Image Person Retrieval
The goal of Text-to-Image Person Retrieval (TIPR) is to retrieve specific person images according to the given textual descriptions. A primary challenge in this task is bridging the substantial representational gap between visual and textual modalities. The prevailing methods map texts and images into unified embedding space for matching, while the intricate semantic correspondences between texts and images are still not effectively constructed. To address this issue, we propose a novel TIPR framework to build fine-grained interactions and alignment between person images and the corresponding texts. Specifically, via fine-tuning the Contrastive Language-Image Pre-training (CLIP) model, a visual-textual dual encoder is firstly constructed, to preliminarily align the image and text features. Secondly, a Text-guided Image Restoration (TIR) auxiliary task is proposed to map abstract textual entities to specific image regions, improving the alignment between local textual and visual embeddings. Additionally, a cross-modal triplet loss is presented to handle hard samples, and further enhance the model's discriminability for minor differences. Moreover, a pruning-based text data augmentation approach is proposed to enhance focus on essential elements in descriptions, thereby avoiding excessive model attention to less significant information. The experimental results show our proposed method outperforms state-of-the-art methods on three popular benchmark datasets, and the code will be made publicly available at https://github.com/Delong-liu-bupt/SEN.
comment: The paper was withdrawn due to a dispute among the authors regarding the content of the article
♻ ☆ Logic Augmented Generation
Semantic Knowledge Graphs (SKG) face challenges with scalability, flexibility, contextual understanding, and handling unstructured or ambiguous information. However, they offer formal and structured knowledge enabling highly interpretable and reliable results by means of reasoning and querying. Large Language Models (LLMs) overcome those limitations making them suitable in open-ended tasks and unstructured environments. Nevertheless, LLMs are neither interpretable nor reliable. To solve the dichotomy between LLMs and SKGs we envision Logic Augmented Generation (LAG) that combines the benefits of the two worlds. LAG uses LLMs as Reactive Continuous Knowledge Graphs that can generate potentially infinite relations and tacit knowledge on-demand. SKGs are key for injecting a discrete heuristic dimension with clear logical and factual boundaries. We exemplify LAG in two tasks of collective intelligence, i.e., medical diagnostics and climate projections. Understanding the properties and limitations of LAG, which are still mostly unknown, is of utmost importance for enabling a variety of tasks involving tacit knowledge in order to provide interpretable and effective results.
comment: 10 pages, 2 figures
♻ ☆ Personalized LLM Response Generation with Parameterized Memory Injection
Large Language Models (LLMs) have exhibited remarkable proficiency in comprehending and generating natural language. On the other hand, personalized LLM response generation holds the potential to offer substantial benefits for individuals in critical areas such as medical. Existing research has explored memory-augmented methods to prompt the LLM with pre-stored user-specific knowledge for personalized response generation in terms of new queries. We contend that such paradigm is unable to perceive fine-granularity information. In this study, we propose a novel \textbf{M}emory-\textbf{i}njected approach using parameter-efficient fine-tuning (PEFT) and along with a Bayesian Optimisation searching strategy to achieve \textbf{L}LM \textbf{P}ersonalization(\textbf{MiLP}).
♻ ☆ WebWalker: Benchmarking LLMs in Web Traversal
Retrieval-augmented generation (RAG) demonstrates remarkable performance across tasks in open-domain question-answering. However, traditional search engines may retrieve shallow content, limiting the ability of LLMs to handle complex, multi-layered information. To address it, we introduce WebWalkerQA, a benchmark designed to assess the ability of LLMs to perform web traversal. It evaluates the capacity of LLMs to traverse a website's subpages to extract high-quality data systematically. We propose WebWalker, which is a multi-agent framework that mimics human-like web navigation through an explore-critic paradigm. Extensive experimental results show that WebWalkerQA is challenging and demonstrates the effectiveness of RAG combined with WebWalker, through the horizontal and vertical integration in real-world scenarios.
♻ ☆ Inductive Learning of Logical Theories with LLMs: An Expressivity-Graded Analysis
This work presents a novel systematic methodology to analyse the capabilities and limitations of Large Language Models (LLMs) with feedback from a formal inference engine, on logic theory induction. The analysis is complexity-graded w.r.t. rule dependency structure, allowing quantification of specific inference challenges on LLM performance. Integrating LLMs with formal methods is a promising frontier in the Natural Language Processing field, as an important avenue for improving model inference control and explainability. In particular, inductive learning over complex sets of facts and rules, poses unique challenges for current autoregressive models, as they lack explicit symbolic grounding. While they can be complemented by formal systems, the properties delivered by LLMs regarding inductive learning, are not well understood and quantified. Empirical results indicate that the largest LLMs can achieve competitive results against a SOTA Inductive Logic Programming (ILP) system baseline, but also that tracking long predicate relationship chains is a more difficult obstacle than theory complexity for LLMs.
♻ ☆ Are LLMs Good Literature Review Writers? Evaluating the Literature Review Writing Ability of Large Language Models
The literature review is a crucial form of academic writing that involves complex processes of literature collection, organization, and summarization. The emergence of large language models (LLMs) has introduced promising tools to automate these processes. However, their actual capabilities in writing comprehensive literature reviews remain underexplored, such as whether they can generate accurate and reliable references. To address this gap, we propose a framework to assess the literature review writing ability of LLMs automatically. We evaluate the performance of LLMs across three tasks: generating references, writing abstracts, and writing literature reviews. We employ external tools for a multidimensional evaluation, which includes assessing hallucination rates in references, semantic coverage, and factual consistency with human-written context. By analyzing the experimental results, we find that, despite advancements, even the most sophisticated models still cannot avoid generating hallucinated references. Additionally, different models exhibit varying performance in literature review writing across different disciplines.
comment: 12 pages, 5 figures, 5 tables
♻ ☆ Addressing Hallucinations in Language Models with Knowledge Graph Embeddings as an Additional Modality
In this paper we present an approach to reduce hallucinations in Large Language Models (LLMs) by incorporating Knowledge Graphs (KGs) as an additional modality. Our method involves transforming input text into a set of KG embeddings and using an adapter to integrate these embeddings into the language model space, without relying on external retrieval processes. To facilitate this, we created WikiEntities, a dataset containing over 3 million Wikipedia texts annotated with entities from Wikidata and their corresponding embeddings from PyTorch-BigGraph. This dataset serves as a valuable resource for training Entity Linking models and adapting the described method to various LLMs using specialized adapters. Our method does not require fine-tuning of the language models themselves; instead, we only train the adapter. This ensures that the model's performance on other tasks is not affected. We trained an adapter for the Mistral 7B, LLaMA 2-7B (chat), and LLaMA 3-8B (instruct) models using this dataset and demonstrated that our approach improves performance on the HaluEval, True-False benchmarks and FEVER dataset. The results indicate that incorporating KGs as a new modality can effectively reduce hallucinations and improve the factual accuracy of language models, all without the need for external retrieval.
♻ ☆ JsonTuning: Towards Generalizable, Robust, and Controllable Instruction Tuning
Instruction tuning is vital for enhancing the performance of large language models (LLMs), but existing text-to-text methods, referred to as TextTuning, struggle with issues such as generalization, robustness, and controllability due to their lack of explicit task structures. We introduce JsonTuning, a structure-to-structure approach that uses JSON structures to represent tasks. This method improves generalization by clarifying task elements and their relations, boosts robustness by minimizing ambiguity, and enhances controllability by allowing precise control over outputs. We conduct an extensive comparative analysis between JsonTuning and TextTuning using various language models and benchmarks. Our findings reveal that JsonTuning consistently surpasses TextTuning in terms of performance, robustness, and controllability across different scenarios. By overcoming the limitations of TextTuning, JsonTuning demonstrates significant potential for developing more effective and reliable LLMs capable of handling diverse scenarios.
♻ ☆ TreeKV: Smooth Key-Value Cache Compression with Tree Structures
Efficient key-value (KV) cache compression is critical for scaling transformer-based Large Language Models (LLMs) in long sequences and resource-limited settings. Existing methods evict tokens based on their positions or importance scores, but position-based strategies can miss crucial information outside predefined regions, while those relying on global importance scores resulting in strong regional biases, limiting the KV cache's overall context retention and potentially impairing the performance of LLMs on complex tasks. Our wavelet analysis reveals that as tokens approach the end of sequence, their contributions to generation gradually increase and tends to diverge more from neighboring tokens, indicating a smooth transition with increasing complexity and variability from distant to nearby context. Motivated by this observation, we propose TreeKV, an intuitive, training-free method that employs a tree structure for smooth cache compression. TreeKV maintains a fixed cache size, allowing LLMs to deliver high-quality output even in long text scenarios. Unlike most compression methods, TreeKV is applicable to both the generation and prefilling stages. TreeKV consistently surpasses all baseline models in language modeling tasks on PG19 and OpenWebText2, allowing LLMs trained with short context window to generalize to longer window with a 16x cache reduction. On the Longbench benchmark, TreeKV achieves the best performance with only 6\% of the budget at optimal efficiency.
♻ ☆ AdaptVC: High Quality Voice Conversion with Adaptive Learning ICASSP 2025
The goal of voice conversion is to transform the speech of a source speaker to sound like that of a reference speaker while preserving the original content. A key challenge is to extract disentangled linguistic content from the source and voice style from the reference. While existing approaches leverage various methods to isolate the two, a generalization still requires further attention, especially for robustness in zero-shot scenarios. In this paper, we achieve successful disentanglement of content and speaker features by tuning self-supervised speech features with adapters. The adapters are trained to dynamically encode nuanced features from rich self-supervised features, and the decoder fuses them to produce speech that accurately resembles the reference with minimal loss of content. Moreover, we leverage a conditional flow matching decoder with cross-attention speaker conditioning to further boost the synthesis quality and efficiency. Subjective and objective evaluations in a zero-shot scenario demonstrate that the proposed method outperforms existing models in speech quality and similarity to the reference speech.
comment: ICASSP 2025; demo available https://mm.kaist.ac.kr/projects/AdaptVC
♻ ☆ Transformers and Large Language Models for Efficient Intrusion Detection Systems: A Comprehensive Survey
With significant advancements in Transformers LLMs, NLP has extended its reach into many research fields due to its enhanced capabilities in text generation and user interaction. One field benefiting greatly from these advancements is cybersecurity. In cybersecurity, many parameters that need to be protected and exchanged between senders and receivers are in the form of text and tabular data, making NLP a valuable tool in enhancing the security measures of communication protocols. This survey paper provides a comprehensive analysis of the utilization of Transformers and LLMs in cyber-threat detection systems. The methodology of paper selection and bibliometric analysis is outlined to establish a rigorous framework for evaluating existing research. The fundamentals of Transformers are discussed, including background information on various cyber-attacks and datasets commonly used in this field. The survey explores the application of Transformers in IDSs, focusing on different architectures such as Attention-based models, LLMs like BERT and GPT, CNN/LSTM-Transformer hybrids, emerging approaches like ViTs, among others. Furthermore, it explores the diverse environments and applications where Transformers and LLMs-based IDS have been implemented, including computer networks, IoT devices, critical infrastructure protection, cloud computing, SDN, as well as in autonomous vehicles. The paper also addresses research challenges and future directions in this area, identifying key issues such as interpretability, scalability, and adaptability to evolving threats, and more. Finally, the conclusion summarizes the findings and highlights the significance of Transformers and LLMs in enhancing cyber-threat detection capabilities, while also outlining potential avenues for further research and development.
comment: arXiv admin note: text overlap with arXiv:2405.04760 by other authors
♻ ☆ One Language, Many Gaps: Evaluating Dialect Fairness and Robustness of Large Language Models in Reasoning Tasks
Language is not monolithic. While benchmarks, including those designed for multiple languages, are often used as proxies to evaluate the performance of Large Language Models (LLMs), they tend to overlook the nuances of within-language variation, and thus fail to model the experience of speakers of non-standard dialects. Focusing on African American Vernacular English (AAVE), we present the first study aimed at objectively assessing the fairness and robustness of LLMs in handling dialects in canonical reasoning tasks, including algorithm, math, logic, and integrated reasoning. We introduce \textbf{ReDial} (\textbf{Re}asoning with \textbf{Dial}ect Queries), a benchmark containing 1.2K+ parallel query pairs in Standardized English and AAVE. We hire AAVE speakers, including experts with computer science backgrounds, to rewrite seven popular benchmarks, such as HumanEval and GSM8K. With ReDial, we evaluate widely used LLMs, including GPT, Claude, Llama, Mistral, and the Phi model families. Our findings reveal that \textbf{almost all of these widely used models show significant brittleness and unfairness to queries in AAVE}. Our work establishes a systematic and objective framework for analyzing LLM bias in dialectal queries. Moreover, it highlights how mainstream LLMs provide unfair service to dialect speakers in reasoning tasks, laying a critical foundation for relevant future research. Code and data can be accessed at https://github.com/fangru-lin/redial_dialect_robustness_fairness.
♻ ☆ MoPE: Mixture of Prompt Experts for Parameter-Efficient and Scalable Multimodal Fusion
Despite the demonstrated parameter efficiency of prompt-based multimodal fusion methods, their limited adaptivity and expressiveness often result in suboptimal performance compared to other tuning approaches. In this paper, we introduce the Mixture of Prompt Experts (MoPE), the first technique designed to overcome these limitations by decomposing standard prompts to capture instance-level features adaptively. Building on this decomposition, MoPE enhances prompt fusion's expressiveness by leveraging multimodal pairing priors to route the most effective prompt for each instance dynamically. Compared to vanilla prompting, our MoPE-based fusion method exhibits greater expressiveness, scaling more effectively with the training data and the overall number of trainable parameters. We also investigate regularization terms for expert routing, which lead to emergent expert specialization with enhanced adaptiveness and interpretablity. Extensive experiments across six multimodal datasets spanning four modalities demonstrate state-of-the-art performance for prompt fusion, matching or even surpassing the performance of fine-tuning while requiring only 0.8% of the trainable parameters. Project homepage: https://github.com/songrise/MoPE
comment: Under Review, Extended version of arxiv:2312.03734
♻ ☆ UTMath: Math Evaluation with Unit Test via Reasoning-to-Coding Thoughts
The evaluation of mathematical reasoning capabilities is essential for advancing Artificial General Intelligence (AGI). While Large Language Models (LLMs) have shown impressive performance in solving mathematical problems, existing benchmarks such as GSM8K and MATH present limitations, including narrow problem definitions with specific numbers and reliance on predetermined rules that hinder accurate assessments of reasoning and generality. This paper introduces the UTMath Benchmark, a robust evaluation framework designed to assess LLMs through extensive unit tests, with a focus on both the accuracy and generality of model responses. It comprises 1,053 cutting-edge problems spanning nine mathematical domains, with an average of 68 test cases per problem. UTMath is highly challenging, with the best-performing model, o1-mini, solving only 32.57\% of the problems, followed by o1-preview at 27.16\%, and GPT-4o at 26.93\%. Furthermore, we present the Reasoning-to-Coding of Thoughts (RCoT) approach, which encourages LLMs to engage in explicit reasoning prior to code generation, thereby facilitating the production of more sophisticated solutions and enhancing overall performance and efficiency. Additionally, we also release the UTMath-Train training dataset (more than 70k samples), to support the community in further exploring mathematical reasoning. Our benchmark can be accessed via the following link: https://github.com/UTMathGroup/UTMath
♻ ☆ BIOMEDICA: An Open Biomedical Image-Caption Archive, Dataset, and Vision-Language Models Derived from Scientific Literature
The development of vision-language models (VLMs) is driven by large-scale and diverse multimodal datasets. However, progress toward generalist biomedical VLMs is limited by the lack of annotated, publicly accessible datasets across biology and medicine. Existing efforts are restricted to narrow domains, missing the full diversity of biomedical knowledge encoded in scientific literature. To address this gap, we introduce BIOMEDICA, a scalable, open-source framework to extract, annotate, and serialize the entirety of the PubMed Central Open Access subset into an easy-to-use, publicly accessible dataset. Our framework produces a comprehensive archive with over 24 million unique image-text pairs from over 6 million articles. Metadata and expert-guided annotations are also provided. We demonstrate the utility and accessibility of our resource by releasing BMCA-CLIP, a suite of CLIP-style models continuously pre-trained on the BIOMEDICA dataset via streaming, eliminating the need to download 27 TB of data locally. On average, our models achieve state-of-the-art performance across 40 tasks - spanning pathology, radiology, ophthalmology, dermatology, surgery, molecular biology, parasitology, and cell biology - excelling in zero-shot classification with a 6.56% average improvement (as high as 29.8% and 17.5% in dermatology and ophthalmology, respectively), and stronger image-text retrieval, all while using 10x less compute. To foster reproducibility and collaboration, we release our codebase and dataset for the broader research community.
♻ ☆ FLM-101B: An Open LLM and How to Train It with $100K Budget
Large language models (LLMs) are considered important approaches towards foundational machine intelligence, achieving remarkable success in Natural Language Processing and multimodal tasks, among others. However, the carbon footprints and financial costs originating from heavy pre-training computation is a non-negligible issue. Progressive training methods, inspired by the neurogenesis process that grows neural structures, have shown potential to accelerate LLM pre-training. However, the algorithms, implementation, and practices for progressively training LLMs beyond 100B parameters remain underexplored. In this paper, we show that our model, namely FLM-101B, trained with our growth strategy under a budget of \$100K, reaches 80\% of the baselines' performances with only 10\% of their floating-point operations. We believe that further studies on progressive training will benefit the community by cutting down the costs and promoting green AI. The checkpoint of FLM-101B is released at https://huggingface.co/CofeAI/FLM-101B.
♻ ☆ What Makes Cryptic Crosswords Challenging for LLMs? COLING 2025
Cryptic crosswords are puzzles that rely on general knowledge and the solver's ability to manipulate language on different levels, dealing with various types of wordplay. Previous research suggests that solving such puzzles is challenging even for modern NLP models, including Large Language Models (LLMs). However, there is little to no research on the reasons for their poor performance on this task. In this paper, we establish the benchmark results for three popular LLMs: Gemma2, LLaMA3 and ChatGPT, showing that their performance on this task is still significantly below that of humans. We also investigate why these models struggle to achieve superior performance. We release our code and introduced datasets at https://github.com/bodasadallah/decrypting-crosswords.
comment: COLING 2025. arXiv admin note: text overlap with arXiv:2403.12094
♻ ☆ Multi-matrix Factorization Attention
We propose novel attention architectures, Multi-matrix Factorization Attention (MFA) and MFA-Key-Reuse (MFA-KR). Existing variants for standard Multi-Head Attention (MHA), including SOTA methods like MLA, fail to maintain as strong performance under stringent Key-Value cache (KV cache) constraints. MFA enhances model capacity by efficiently scaling up both the number and dimension of attention heads through low-rank matrix factorization in the Query-Key (QK) circuit. Extending MFA, MFA-KR further reduces memory requirements by repurposing the key cache as value through value projection re-parameterization. MFA's design enables strong model capacity when working under tight KV cache budget, while MFA-KR is suitable for even harsher KV cache limits with minor performance trade-off. Notably, in our extensive and large-scale experiments, the proposed architecture outperforms MLA and performs comparably to MHA, while reducing KV cache usage by up to 56% and 93.7%, respectively.
♻ ☆ Evaluating Mathematical Reasoning Beyond Accuracy AAAI 2025
The leaderboard of Large Language Models (LLMs) in mathematical tasks has been continuously updated. However, the majority of evaluations focus solely on the final results, neglecting the quality of the intermediate steps. This oversight can mask underlying problems, such as logical errors or unnecessary steps in the reasoning process. To measure reasoning beyond final-answer accuracy, we introduce ReasonEval, a new methodology for evaluating the quality of reasoning steps. ReasonEval employs validity and redundancy to characterize the reasoning quality, as well as accompanying LLMs to assess them automatically. We explore different design options for the LLM-based evaluators and empirically demonstrate that ReasonEval, when instantiated with base models possessing strong mathematical knowledge and trained with high-quality labeled data, consistently outperforms baseline methods in the meta-evaluation datasets. We also highlight the strong generalization capabilities of ReasonEval. By utilizing ReasonEval to evaluate LLMs specialized in math, we find that an increase in final-answer accuracy does not necessarily guarantee an improvement in the overall quality of the reasoning steps for challenging mathematical problems. Additionally, we observe that ReasonEval can play a significant role in data selection. We open-source the best-performing model, meta-evaluation script, and all evaluation results to facilitate future research.
comment: v2 is the AAAI 2025 camera ready version. Project site with code: https://github.com/GAIR-NLP/ReasonEval
♻ ☆ AfriMed-QA: A Pan-African, Multi-Specialty, Medical Question-Answering Benchmark Dataset
Recent advancements in large language model(LLM) performance on medical multiple choice question (MCQ) benchmarks have stimulated interest from healthcare providers and patients globally. Particularly in low-and middle-income countries (LMICs) facing acute physician shortages and lack of specialists, LLMs offer a potentially scalable pathway to enhance healthcare access and reduce costs. However, their effectiveness in the Global South, especially across the African continent, remains to be established. In this work, we introduce AfriMed-QA, the first large scale Pan-African English multi-specialty medical Question-Answering (QA) dataset, 15,000 questions (open and closed-ended) sourced from over 60 medical schools across 16 countries, covering 32 medical specialties. We further evaluate 30 LLMs across multiple axes including correctness and demographic bias. Our findings show significant performance variation across specialties and geographies, MCQ performance clearly lags USMLE (MedQA). We find that biomedical LLMs underperform general models and smaller edge-friendly LLMs struggle to achieve a passing score. Interestingly, human evaluations show a consistent consumer preference for LLM answers and explanations when compared with clinician answers.
♻ ☆ Joint Beam Search Integrating CTC, Attention, and Transducer Decoders
End-to-end automatic speech recognition (E2E-ASR) can be classified by its decoder architectures, such as connectionist temporal classification (CTC), recurrent neural network transducer (RNN-T), attention-based encoder-decoder, and Mask-CTC models. Each decoder architecture has advantages and disadvantages, leading practitioners to switch between these different models depending on application requirements. Instead of building separate models, we propose a joint modeling scheme where four decoders (CTC, RNN-T, attention, and Mask-CTC) share the same encoder -- we refer to this as 4D modeling. The 4D model is trained jointly, which will bring model regularization and maximize the model robustness thanks to their complementary properties. To efficiently train the 4D model, we introduce a two-stage training strategy that stabilizes the joint training. In addition, we propose three novel joint beam search algorithms by combining three decoders (CTC, RNN-T, and attention) to further improve performance. These three beam search algorithms differ in which decoder is used as the primary decoder. We carefully evaluate the performance and computational tradeoffs associated with each algorithm. Experimental results demonstrate that the jointly trained 4D model outperforms the E2E-ASR models trained with only one individual decoder. Furthermore, we demonstrate that the proposed joint beam search algorithm outperforms the previously proposed CTC/attention decoding.
comment: accepted to IEEE/ACM Transactions on Audio Speech and Language Processing
♻ ☆ ELDER: Enhancing Lifelong Model Editing with Mixture-of-LoRA AAAI-25
Large language models (LLMs) require model editing to efficiently update specific knowledge within them and avoid factual errors. Most model editing methods are solely designed for single-time use and result in a significant forgetting effect in lifelong editing scenarios, where sequential edits are conducted over time. Previous approaches manage sequential edits by freezing original parameters and discretely allocating new parameters for each knowledge update. However, these methods lack robustness to minor input variations due to the discrete mapping between data and parameters. To overcome this challenge, we propose ELDER, a novel approach to create a continuous association between data and adapters. ELDER integrates multiple LoRAs through a router network and is trained to establish a smooth data-adapter association, thereby enhancing the edit robustness and generalization of semantically equivalent inputs. To ensure inputs containing the same knowledge will be processed by the same LoRAs, we design a novel loss to guide the model link LoRA allocations with edit knowledge. Furthermore, we propose a deferral mechanism to retain the original LLM capabilities post-edit. Extensive experiments on GPT-2 XL and LLaMA2-7B demonstrate that ELDER effectively edits models in the lifelong setting, outperforming eight baselines while exhibiting strong scalability and preserving LLMs' general abilities on downstream tasks. Our code is available at https://github.com/JiaangL/ELDER.
comment: Accepted by AAAI-25
♻ ☆ Retrieval-Reasoning Large Language Model-based Synthetic Clinical Trial Generation
Machine learning (ML) exhibits promise in the clinical domain. However, it is constrained by data scarcity and ethical considerations, as the generation of clinical trials presents significant challenges due to stringent privacy regulations, high costs, and the extended duration required for conducting studies with human participants. Despite the advancements of large language models (LLMs) in general generation tasks, their potential in facilitating the generation of synthetic clinical trials is under-explored. To address this gap, we introduce a novel Retrieval-Reasoning few-shot framework that leverages LLMs to generate artificial yet realistic and diverse clinical trials with binary success/failure labels. Experiments conducted on real clinical trials from the \url{ClinicalTrials.gov} database demonstrate that our synthetic data can effectively augment real datasets. Furthermore, by fine-tuning a pre-trained model as a binary classifier on synthetic clinical trial datasets, we demonstrate that this augmentation enhances model training for downstream tasks such as trial outcome prediction. Our findings suggest that LLMs for synthetic clinical trial generation hold promise for accelerating clinical research and upholding ethical standards for patient privacy. The code is publicly available at https://anonymous.4open.science/r/Retrieval_Reasoning_Clinical_Trial_Generation-3EC4.
♻ ☆ Energy-Efficient Split Learning for Fine-Tuning Large Language Models in Edge Networks
In this letter, we propose an energy-efficient split learning (SL) framework for fine-tuning large language models (LLMs) using geo-distributed personal data at the network edge, where LLMs are split and alternately across massive mobile devices and an edge server. Considering the device heterogeneity and channel dynamics in edge networks, a \underline{C}ut l\underline{A}yer and computing \underline{R}esource \underline{D}ecision (CARD) algorithm is developed to minimize training delay and energy consumption. Simulation results demonstrate that the proposed approach reduces the average training delay and server's energy consumption by 70.8% and 53.1%, compared to the benchmarks, respectively.
comment: 5 pages, 6 figures
♻ ☆ Don't Command, Cultivate: An Exploratory Study of System-2 Alignment
The o1 system card identifies the o1 models as the most robust within OpenAI, with their defining characteristic being the progression from rapid, intuitive thinking to slower, more deliberate reasoning. This observation motivated us to investigate the influence of System-2 thinking patterns on model safety. In our preliminary research, we conducted safety evaluations of the o1 model, including complex jailbreak attack scenarios using adversarial natural language prompts and mathematical encoding prompts. Our findings indicate that the o1 model demonstrates relatively improved safety performance; however, it still exhibits vulnerabilities, particularly against jailbreak attacks employing mathematical encoding. Through detailed case analysis, we identified specific patterns in the o1 model's responses. We also explored the alignment of System-2 safety in open-source models using prompt engineering and supervised fine-tuning techniques. Experimental results show that some simple methods to encourage the model to carefully scrutinize user requests are beneficial for model safety. Additionally, we proposed a implementation plan for process supervision to enhance safety alignment. The implementation details and experimental results will be provided in future versions.
comment: In this version, the DPO and reinforcement learning methods have been added
♻ ☆ $\text{Transformer}^2$: Self-adaptive LLMs
Self-adaptive large language models (LLMs) aim to solve the challenges posed by traditional fine-tuning methods, which are often computationally intensive and static in their ability to handle diverse tasks. We introduce $\text{Transformer}^2$, a novel self-adaptation framework that adapts LLMs for unseen tasks in real-time by selectively adjusting only the singular components of their weight matrices. During inference, $\text{Transformer}^2$ employs a two-pass mechanism: first, a dispatch system identifies the task properties, and then task-specific "expert" vectors, trained using reinforcement learning, are dynamically mixed to obtain targeted behavior for the incoming prompt. Our method outperforms ubiquitous approaches such as LoRA, with fewer parameters and greater efficiency. $\text{Transformer}^2$ demonstrates versatility across different LLM architectures and modalities, including vision-language tasks. $\text{Transformer}^2$ represents a significant leap forward, offering a scalable, efficient solution for enhancing the adaptability and task-specific performance of LLMs, paving the way for truly dynamic, self-organizing AI systems.
comment: 18 panges, 11 figures, 9 tables
♻ ☆ Gradient descent with generalized Newton's method
We propose the generalized Newton's method (GeN) -- a Hessian-informed approach that applies to any optimizer such as SGD and Adam, and covers the Newton-Raphson method as a sub-case. Our method automatically and dynamically selects the learning rate that accelerates the convergence, without the intensive tuning of the learning rate scheduler. In practice, our method is easily implementable, since it only requires additional forward passes with almost zero computational overhead (in terms of training time and memory cost), if the overhead is amortized over many iterations. We present extensive experiments on language and vision tasks (e.g. GPT and ResNet) to showcase that GeN optimizers match the state-of-the-art performance, which was achieved with carefully tuned learning rate schedulers.
♻ ☆ Let the Rule Speak: Enhancing In-context Learning Debiasing with Interpretability
In-context learning, which allows large language models to perform diverse tasks with a few demonstrations, is found to have imbalanced per-class prediction accuracy on multi-class text classification. Although notable output correction methods have been developed to tackle the issue and simultaneously improve downstream prediction accuracy, they may fail to answer the core interpretability challenges: why and which certain classes need corrections, and more importantly, a tailored correction for per-sample, per-class's probability. To address such interpretability gaps, we first find that the imbalance arises from certain classes consistently receiving high ICL output probabilities, whereas others receiving lower or mixed ranges, so the former is more frequently chosen, resulting in higher accuracy; more crucially, we find that these ranges have significantly varying degrees of influence on the accuracy bias, highlighting the need for precise, interpretable probability corrections by range. Motivated by this, we propose FuRud, a Fuzzy Rule Optimization based Debiasing method, that (1) detects which classes need corrections, and (2) for each correction-needed class, detects its probability ranges and applies asymmetric amplifications or reductions to correct them interpretably. Notably, across seven benchmark datasets, FuRud reduces the pairwise class accuracy bias (COBias) by more than half (56%), while achieving a relative increase of 21% in accuracy, outperforming state-of-the-art debiasing methods. Moreover, FuRud can optimize downstream tasks with as few as 10 optimization examples. Furthermore, FuRud can work for prompt formats that lead to highly skewed predictions. For example, FuRud greatly improves ICL outputs which use letter options, with 44% relative accuracy increase and 54% relative COBias reduction.
♻ ☆ Double Equivariance for Inductive Link Prediction for Both New Nodes and New Relation Types
The task of fully inductive link prediction in knowledge graphs has gained significant attention, with various graph neural networks being proposed to address it. This task presents greater challenges than traditional inductive link prediction tasks with only new nodes, as models must be capable of zero-shot generalization to both unseen nodes and unseen relation types in the inference graph. Despite the development of novel models, a unifying theoretical understanding of their success remains elusive, and the limitations of these methods are not well-studied. In this work, we introduce the concept of double permutation-equivariant representations and demonstrate its necessity for effective performance in this task. We show that many existing models, despite their diverse architectural designs, conform to this framework. However, we also identify inherent limitations in double permutation-equivariant representations, which restrict these models's ability to learn effectively on datasets with varying characteristics. Our findings suggest that while double equivariance is necessary for meta-learning across knowledge graphs from different domains, it is not sufficient. There remains a fundamental gap between double permutation-equivariant models and the concept of foundation models designed to learn patterns across all domains.
♻ ☆ LLM Reading Tea Leaves: Automatically Evaluating Topic Models with Large Language Models ACL
Topic modeling has been a widely used tool for unsupervised text analysis. However, comprehensive evaluations of a topic model remain challenging. Existing evaluation methods are either less comparable across different models (e.g., perplexity) or focus on only one specific aspect of a model (e.g., topic quality or document representation quality) at a time, which is insufficient to reflect the overall model performance. In this paper, we propose WALM (Word Agreement with Language Model), a new evaluation method for topic modeling that considers the semantic quality of document representations and topics in a joint manner, leveraging the power of Large Language Models (LLMs). With extensive experiments involving different types of topic models, WALM is shown to align with human judgment and can serve as a complementary evaluation method to the existing ones, bringing a new perspective to topic modeling. Our software package is available at https://github.com/Xiaohao-Yang/Topic_Model_Evaluation.
comment: Forthcoming in Transactions of the Association for Computational Linguistics (TACL) published by MIT Press
Machine Learning 160
☆ Gradient Equilibrium in Online Learning: Theory and Applications
We present a new perspective on online learning that we refer to as gradient equilibrium: a sequence of iterates achieves gradient equilibrium if the average of gradients of losses along the sequence converges to zero. In general, this condition is not implied by nor implies sublinear regret. It turns out that gradient equilibrium is achievable by standard online learning methods such as gradient descent and mirror descent with constant step sizes (rather than decaying step sizes, as is usually required for no regret). Further, as we show through examples, gradient equilibrium translates into an interpretable and meaningful property in online prediction problems spanning regression, classification, quantile estimation, and others. Notably, we show that the gradient equilibrium framework can be used to develop a debiasing scheme for black-box predictions under arbitrary distribution shift, based on simple post hoc online descent updates. We also show that post hoc gradient updates can be used to calibrate predicted quantiles under distribution shift, and that the framework leads to unbiased Elo scores for pairwise preference prediction.
comment: Code available at https://github.com/aangelopoulos/gradient-equilibrium/
☆ A Similarity Measure Between Functions with Applications to Statistical Learning and Optimization
In this note, we present a novel measure of similarity between two functions. It quantifies how the sub-optimality gaps of two functions convert to each other, and unifies several existing notions of functional similarity. We show that it has convenient operation rules, and illustrate its use in empirical risk minimization and non-stationary online optimization.
comment: 9 pages
☆ Diffusion Adversarial Post-Training for One-Step Video Generation
The diffusion models are widely used for image and video generation, but their iterative generation process is slow and expansive. While existing distillation approaches have demonstrated the potential for one-step generation in the image domain, they still suffer from significant quality degradation. In this work, we propose Adversarial Post-Training (APT) against real data following diffusion pre-training for one-step video generation. To improve the training stability and quality, we introduce several improvements to the model architecture and training procedures, along with an approximated R1 regularization objective. Empirically, our experiments show that our adversarial post-trained model, Seaweed-APT, can generate 2-second, 1280x720, 24fps videos in real time using a single forward evaluation step. Additionally, our model is capable of generating 1024px images in a single step, achieving quality comparable to state-of-the-art methods.
☆ Path Loss Prediction Using Machine Learning with Extended Features
Wireless communications rely on path loss modeling, which is most effective when it includes the physical details of the propagation environment. Acquiring this data has historically been challenging, but geographic information system data is becoming increasingly available with higher resolution and accuracy. Access to such details enables propagation models to more accurately predict coverage and minimize interference in wireless deployments. Machine learning-based modeling can significantly support this effort, with feature-based approaches allowing for accurate, efficient, and scalable propagation modeling. Building on previous work, we introduce an extended set of features that improves prediction accuracy while, most importantly, maintaining model generalization across a broad range of environments.
comment: 4 pages, 4 figures, conference paper
☆ Benchmarking Graph Representations and Graph Neural Networks for Multivariate Time Series Classification
Multivariate Time Series Classification (MTSC) enables the analysis if complex temporal data, and thus serves as a cornerstone in various real-world applications, ranging from healthcare to finance. Since the relationship among variables in MTS usually contain crucial cues, a large number of graph-based MTSC approaches have been proposed, as the graph topology and edges can explicitly represent relationships among variables (channels), where not only various MTS graph representation learning strategies but also different Graph Neural Networks (GNNs) have been explored. Despite such progresses, there is no comprehensive study that fairly benchmarks and investigates the performances of existing widely-used graph representation learning strategies/GNN classifiers in the application of different MTSC tasks. In this paper, we present the first benchmark which systematically investigates the effectiveness of the widely-used three node feature definition strategies, four edge feature learning strategies and five GNN architecture, resulting in 60 different variants for graph-based MTSC. These variants are developed and evaluated with a standardized data pipeline and training/validation/testing strategy on 26 widely-used suspensor MTSC datasets. Our experiments highlight that node features significantly influence MTSC performance, while the visualization of edge features illustrates why adaptive edge learning outperforms other edge feature learning methods. The code of the proposed benchmark is publicly available at \url{https://github.com/CVI-yangwn/Benchmark-GNN-for-Multivariate-Time-Series-Classification}.
☆ Polynomial Threshold Functions of Bounded Tree-Width: Some Explainability and Complexity Aspects
The tree-width of a multivariate polynomial is the tree-width of the hypergraph with hyperedges corresponding to its terms. Multivariate polynomials of bounded tree-width have been studied by Makowsky and Meer as a new sparsity condition that allows for polynomial solvability of problems which are intractable in general. We consider a variation on this theme for Boolean variables. A representation of a Boolean function as the sign of a polynomial is called a polynomial threshold representation. We discuss Boolean functions representable as polynomial threshold functions of bounded tree-width and present two applications to Bayesian network classifiers, a probabilistic graphical model. Both applications are in Explainable Artificial Intelligence (XAI), the research area dealing with the black-box nature of many recent machine learning models. We also give a separation result between the representational power of positive and general polynomial threshold functions.
comment: 22 pages, 3 figures. To be published in Festschrift in honor of Johann A. Makowsky
☆ Avoiding subtraction and division of stochastic signals using normalizing flows: NFdeconvolve
Across the scientific realm, we find ourselves subtracting or dividing stochastic signals. For instance, consider a stochastic realization, $x$, generated from the addition or multiplication of two stochastic signals $a$ and $b$, namely $x=a+b$ or $x = ab$. For the $x=a+b$ example, $a$ can be fluorescence background and $b$ the signal of interest whose statistics are to be learned from the measured $x$. Similarly, when writing $x=ab$, $a$ can be thought of as the illumination intensity and $b$ the density of fluorescent molecules of interest. Yet dividing or subtracting stochastic signals amplifies noise, and we ask instead whether, using the statistics of $a$ and the measurement of $x$ as input, we can recover the statistics of $b$. Here, we show how normalizing flows can generate an approximation of the probability distribution over $b$, thereby avoiding subtraction or division altogether. This method is implemented in our software package, NFdeconvolve, available on GitHub with a tutorial linked in the main text.
☆ Can Bayesian Neural Networks Explicitly Model Input Uncertainty?
Inputs to machine learning models can have associated noise or uncertainties, but they are often ignored and not modelled. It is unknown if Bayesian Neural Networks and their approximations are able to consider uncertainty in their inputs. In this paper we build a two input Bayesian Neural Network (mean and standard deviation) and evaluate its capabilities for input uncertainty estimation across different methods like Ensembles, MC-Dropout, and Flipout. Our results indicate that only some uncertainty estimation methods for approximate Bayesian NNs can model input uncertainty, in particular Ensembles and Flipout.
comment: 12 pages, 11 figures, VISAPP 2025 camera ready
☆ Decoding Interpretable Logic Rules from Neural Networks
As deep neural networks continue to excel across various domains, their black-box nature has raised concerns about transparency and trust. In particular, interpretability has become increasingly essential for applications that demand high safety and knowledge rigor, such as drug discovery, autonomous driving, and genomics. However, progress in understanding even the simplest deep neural networks - such as fully connected networks - has been limited, despite their role as foundational elements in state-of-the-art models like ResNet and Transformer. In this paper, we address this challenge by introducing NeuroLogic, a novel approach for decoding interpretable logic rules from neural networks. NeuroLogic leverages neural activation patterns to capture the model's critical decision-making processes, translating them into logical rules represented by hidden predicates. Thanks to its flexible design in the grounding phase, NeuroLogic can be adapted to a wide range of neural networks. For simple fully connected neural networks, hidden predicates can be grounded in certain split patterns of original input features to derive decision-tree-like rules. For large, complex vision neural networks, NeuroLogic grounds hidden predicates into high-level visual concepts that are understandable to humans. Our empirical study demonstrates that NeuroLogic can extract global and interpretable rules from state-of-the-art models such as ResNet, a task at which existing work struggles. We believe NeuroLogic can help pave the way for understanding the black-box nature of neural networks.
comment: 23 pages, 7 figures
☆ AI Driven Water Segmentation with deep learning models for Enhanced Flood Monitoring
Flooding is a major natural hazard causing significant fatalities and economic losses annually, with increasing frequency due to climate change. Rapid and accurate flood detection and monitoring are crucial for mitigating these impacts. This study compares the performance of three deep learning models UNet, ResNet, and DeepLabv3 for pixelwise water segmentation to aid in flood detection, utilizing images from drones, in field observations, and social media. This study involves creating a new dataset that augments wellknown benchmark datasets with flood-specific images, enhancing the robustness of the models. The UNet, ResNet, and DeepLab v3 architectures are tested to determine their effectiveness in various environmental conditions and geographical locations, and the strengths and limitations of each model are also discussed here, providing insights into their applicability in different scenarios by predicting image segmentation masks. This fully automated approach allows these models to isolate flooded areas in images, significantly reducing processing time compared to traditional semi-automated methods. The outcome of this study is to predict segmented masks for each image effected by a flood disaster and the validation accuracy of these models. This methodology facilitates timely and continuous flood monitoring, providing vital data for emergency response teams to reduce loss of life and economic damages. It offers a significant reduction in the time required to generate flood maps, cutting down the manual processing time. Additionally, we present avenues for future research, including the integration of multimodal data sources and the development of robust deep learning architectures tailored specifically for flood detection tasks. Overall, our work contributes to the advancement of flood management strategies through innovative use of deep learning technologies.
comment: 8 pages, 6 figures
☆ Multiplayer Federated Learning: Reaching Equilibrium with Less Communication
Traditional Federated Learning (FL) approaches assume collaborative clients with aligned objectives working towards a shared global model. However, in many real-world scenarios, clients act as rational players with individual objectives and strategic behaviors, a concept that existing FL frameworks are not equipped to adequately address. To bridge this gap, we introduce Multiplayer Federated Learning (MpFL), a novel framework that models the clients in the FL environment as players in a game-theoretic context, aiming to reach an equilibrium. In this scenario, each player tries to optimize their own utility function, which may not align with the collective goal. Within MpFL, we propose Per-Player Local Stochastic Gradient Descent (PEARL-SGD), an algorithm in which each player/client performs local updates independently and periodically communicates with other players. We theoretically analyze PEARL-SGD and prove that it reaches a neighborhood of equilibrium with less communication in the stochastic setup compared to its non-local counterpart. Finally, we verify our theoretical findings through numerical experiments.
comment: 43 pages, 5 figures
☆ FDPP: Fine-tune Diffusion Policy with Human Preference
Imitation learning from human demonstrations enables robots to perform complex manipulation tasks and has recently witnessed huge success. However, these techniques often struggle to adapt behavior to new preferences or changes in the environment. To address these limitations, we propose Fine-tuning Diffusion Policy with Human Preference (FDPP). FDPP learns a reward function through preference-based learning. This reward is then used to fine-tune the pre-trained policy with reinforcement learning (RL), resulting in alignment of pre-trained policy with new human preferences while still solving the original task. Our experiments across various robotic tasks and preferences demonstrate that FDPP effectively customizes policy behavior without compromising performance. Additionally, we show that incorporating Kullback-Leibler (KL) regularization during fine-tuning prevents over-fitting and helps maintain the competencies of the initial policy.
☆ Eliciting In-context Retrieval and Reasoning for Long-context Large Language Models
Recent advancements in long-context language models (LCLMs) promise to transform Retrieval-Augmented Generation (RAG) by simplifying pipelines. With their expanded context windows, LCLMs can process entire knowledge bases and perform retrieval and reasoning directly -- a capability we define as In-Context Retrieval and Reasoning (ICR^2). However, existing benchmarks like LOFT often overestimate LCLM performance by providing overly simplified contexts. To address this, we introduce ICR^2, a benchmark that evaluates LCLMs in more realistic scenarios by including confounding passages retrieved with strong retrievers. We then propose three methods to enhance LCLM performance: (1) retrieve-then-generate fine-tuning, (2) retrieval-attention-probing, which uses attention heads to filter and de-noise long contexts during decoding, and (3) joint retrieval head training alongside the generation head. Our evaluation of five well-known LCLMs on LOFT and ICR^2 demonstrates significant gains with our best approach applied to Mistral-7B: +17 and +15 points by Exact Match on LOFT, and +13 and +2 points on ICR^2, compared to vanilla RAG and supervised fine-tuning, respectively. It even outperforms GPT-4-Turbo on most tasks despite being a much smaller model.
☆ Text-Diffusion Red-Teaming of Large Language Models: Unveiling Harmful Behaviors with Proximity Constraints AAAI 25
Recent work has proposed automated red-teaming methods for testing the vulnerabilities of a given target large language model (LLM). These methods use red-teaming LLMs to uncover inputs that induce harmful behavior in a target LLM. In this paper, we study red-teaming strategies that enable a targeted security assessment. We propose an optimization framework for red-teaming with proximity constraints, where the discovered prompts must be similar to reference prompts from a given dataset. This dataset serves as a template for the discovered prompts, anchoring the search for test-cases to specific topics, writing styles, or types of harmful behavior. We show that established auto-regressive model architectures do not perform well in this setting. We therefore introduce a black-box red-teaming method inspired by text-diffusion models: Diffusion for Auditing and Red-Teaming (DART). DART modifies the reference prompt by perturbing it in the embedding space, directly controlling the amount of change introduced. We systematically evaluate our method by comparing its effectiveness with established methods based on model fine-tuning and zero- and few-shot prompting. Our results show that DART is significantly more effective at discovering harmful inputs in close proximity to the reference prompt.
comment: This is an extended version of a paper published at AAAI 25
☆ Continual Deep Active Learning for Medical Imaging: Replay-Base Architecture for Context Adaptation
Deep Learning for medical imaging faces challenges in adapting and generalizing to new contexts. Additionally, it often lacks sufficient labeled data for specific tasks requiring significant annotation effort. Continual Learning (CL) tackles adaptability and generalizability by enabling lifelong learning from a data stream while mitigating forgetting of previously learned knowledge. Active Learning (AL) reduces the number of required annotations for effective training. This work explores both approaches (CAL) to develop a novel framework for robust medical image analysis. Based on the automatic recognition of shifts in image characteristics, Replay-Base Architecture for Context Adaptation (RBACA) employs a CL rehearsal method to continually learn from diverse contexts, and an AL component to select the most informative instances for annotation. A novel approach to evaluate CAL methods is established using a defined metric denominated IL-Score, which allows for the simultaneous assessment of transfer learning, forgetting, and final model performance. We show that RBACA works in domain and class-incremental learning scenarios, by assessing its IL-Score on the segmentation and diagnosis of cardiac images. The results show that RBACA outperforms a baseline framework without CAL, and a state-of-the-art CAL method across various memory sizes and annotation budgets. Our code is available in https://github.com/RuiDaniel/RBACA .
☆ Engineering LLM Powered Multi-agent Framework for Autonomous CloudOps ICSE 2025
Cloud Operations (CloudOps) is a rapidly growing field focused on the automated management and optimization of cloud infrastructure which is essential for organizations navigating increasingly complex cloud environments. MontyCloud Inc. is one of the major companies in the CloudOps domain that leverages autonomous bots to manage cloud compliance, security, and continuous operations. To make the platform more accessible and effective to the customers, we leveraged the use of GenAI. Developing a GenAI-based solution for autonomous CloudOps for the existing MontyCloud system presented us with various challenges such as i) diverse data sources; ii) orchestration of multiple processes; and iii) handling complex workflows to automate routine tasks. To this end, we developed MOYA, a multi-agent framework that leverages GenAI and balances autonomy with the necessary human control. This framework integrates various internal and external systems and is optimized for factors like task orchestration, security, and error mitigation while producing accurate, reliable, and relevant insights by utilizing Retrieval Augmented Generation (RAG). Evaluations of our multi-agent system with the help of practitioners as well as using automated checks demonstrate enhanced accuracy, responsiveness, and effectiveness over non-agentic approaches across complex workflows.
comment: The paper has been accepted as full paper to CAIN 2025 (https://conf.researchr.org/home/cain-2025), co-located with ICSE 2025 (https://conf.researchr.org/home/icse-2025). The paper was submitted to CAIN for review on 9 November 2024
☆ A Feature-Level Ensemble Model for COVID-19 Identification in CXR Images using Choquet Integral and Differential Evolution Optimization
The COVID-19 pandemic has profoundly impacted billions globally. It challenges public health and healthcare systems due to its rapid spread and severe respiratory effects. An effective strategy to mitigate the COVID-19 pandemic involves integrating testing to identify infected individuals. While RT-PCR is considered the gold standard for diagnosing COVID-19, it has some limitations such as the risk of false negatives. To address this problem, this paper introduces a novel Deep Learning Diagnosis System that integrates pre-trained Deep Convolutional Neural Networks (DCNNs) within an ensemble learning framework to achieve precise identification of COVID-19 cases from Chest X-ray (CXR) images. We combine feature vectors from the final hidden layers of pre-trained DCNNs using the Choquet integral to capture interactions between different DCNNs that a linear approach cannot. We employed Sugeno-$\lambda$ measure theory to derive fuzzy measures for subsets of networks to enable aggregation. We utilized Differential Evolution to estimate fuzzy densities. We developed a TensorFlow-based layer for Choquet operation to facilitate efficient aggregation, due to the intricacies involved in aggregating feature vectors. Experimental results on the COVIDx dataset show that our ensemble model achieved 98\% accuracy in three-class classification and 99.50\% in binary classification, outperforming its components-DenseNet-201 (97\% for three-class, 98.75\% for binary), Inception-v3 (96.25\% for three-class, 98.50\% for binary), and Xception (94.50\% for three-class, 98\% for binary)-and surpassing many previous methods.
☆ Privacy-Preserving Model and Preprocessing Verification for Machine Learning
This paper presents a framework for privacy-preserving verification of machine learning models, focusing on models trained on sensitive data. Integrating Local Differential Privacy (LDP) with model explanations from LIME and SHAP, our framework enables robust verification without compromising individual privacy. It addresses two key tasks: binary classification, to verify if a target model was trained correctly by applying the appropriate preprocessing steps, and multi-class classification, to identify specific preprocessing errors. Evaluations on three real-world datasets-Diabetes, Adult, and Student Record-demonstrate that while the ML-based approach is particularly effective in binary tasks, the threshold-based method performs comparably in multi-class tasks. Results indicate that although verification accuracy varies across datasets and noise levels, the framework provides effective detection of preprocessing errors, strong privacy guarantees, and practical applicability for safeguarding sensitive data.
☆ Dynamic Pricing in High-Speed Railways Using Multi-Agent Reinforcement Learning
This paper addresses a critical challenge in the high-speed passenger railway industry: designing effective dynamic pricing strategies in the context of competing and cooperating operators. To address this, a multi-agent reinforcement learning (MARL) framework based on a non-zero-sum Markov game is proposed, incorporating random utility models to capture passenger decision making. Unlike prior studies in areas such as energy, airlines, and mobile networks, dynamic pricing for railway systems using deep reinforcement learning has received limited attention. A key contribution of this paper is a parametrisable and versatile reinforcement learning simulator designed to model a variety of railway network configurations and demand patterns while enabling realistic, microscopic modelling of user behaviour, called RailPricing-RL. This environment supports the proposed MARL framework, which models heterogeneous agents competing to maximise individual profits while fostering cooperative behaviour to synchronise connecting services. Experimental results validate the framework, demonstrating how user preferences affect MARL performance and how pricing policies influence passenger choices, utility, and overall system dynamics. This study provides a foundation for advancing dynamic pricing strategies in railway systems, aligning profitability with system-wide efficiency, and supporting future research on optimising pricing policies.
comment: 37 pages, 5 figures
☆ Efficient Deep Learning-based Forward Solvers for Brain Tumor Growth Models
Glioblastoma, a highly aggressive brain tumor, poses major challenges due to its poor prognosis and high morbidity rates. Partial differential equation-based models offer promising potential to enhance therapeutic outcomes by simulating patient-specific tumor behavior for improved radiotherapy planning. However, model calibration remains a bottleneck due to the high computational demands of optimization methods like Monte Carlo sampling and evolutionary algorithms. To address this, we recently introduced an approach leveraging a neural forward solver with gradient-based optimization to significantly reduce calibration time. This approach requires a highly accurate and fully differentiable forward model. We investigate multiple architectures, including (i) an enhanced TumorSurrogate, (ii) a modified nnU-Net, and (iii) a 3D Vision Transformer (ViT). The optimized TumorSurrogate achieved the best overall results, excelling in both tumor outline matching and voxel-level prediction of tumor cell concentration. It halved the MSE relative to the baseline model and achieved the highest Dice score across all tumor cell concentration thresholds. Our study demonstrates significant enhancement in forward solver performance and outlines important future research directions.
☆ Big Batch Bayesian Active Learning by Considering Predictive Probabilities NeurIPS
We observe that BatchBALD, a popular acquisition function for batch Bayesian active learning for classification, can conflate epistemic and aleatoric uncertainty, leading to suboptimal performance. Motivated by this observation, we propose to focus on the predictive probabilities, which only exhibit epistemic uncertainty. The result is an acquisition function that not only performs better, but is also faster to evaluate, allowing for larger batches than before.
comment: 7 pages, 2 figures; presented as a lightning talk at the NeurIPS Workshop on Bayesian Decision-making and Uncertainty (BDU; 2024)
☆ Investigating Energy Efficiency and Performance Trade-offs in LLM Inference Across Tasks and DVFS Settings
Large language models (LLMs) have shown significant improvements in many natural language processing (NLP) tasks, accelerating their rapid adoption across many industries. These models are resource-intensive, requiring extensive computational resources both during training and inference, leading to increased energy consumption and negative environmental impact. As their adoption accelerates, the sustainability of LLMs has become a critical issue, necessitating strategies to optimize their runtime efficiency without compromising performance. Hence, it is imperative to identify the parameters that significantly influence the performance and energy efficiency of LLMs. To that end, in this work, we investigate the effect of important parameters on the performance and energy efficiency of LLMs during inference and examine their trade-offs. First, we analyze how different types of models with varying numbers of parameters and architectures perform on tasks like text generation, question answering, and summarization by benchmarking LLMs such as Falcon-7B, Mistral-7B-v0.1, T5-3B, GPT-2, GPT-J-6B, and GPT-Neo-2.7B. Second, we study input and output sequence characteristics such as sequence length concerning energy consumption, performance, and throughput. Finally, we explore the impact of hardware-based power-saving techniques, i.e., Dynamic Voltage Frequency Scaling (DVFS), on the models' latency and energy efficiency. Our extensive benchmarking and statistical analysis reveal many interesting findings, uncovering how specific optimizations can reduce energy consumption while maintaining throughput and accuracy. This study provides actionable insights for researchers and practitioners to design energy-efficient LLM inference systems.
☆ Modeling Feature Maps for Quantum Machine Learning
Quantum Machine Learning (QML) offers significant potential for complex tasks like genome sequence classification, but quantum noise on Noisy Intermediate-Scale Quantum (NISQ) devices poses practical challenges. This study systematically evaluates how various quantum noise models including dephasing, amplitude damping, depolarizing, thermal noise, bit-flip, and phase-flip affect key QML algorithms (QSVC, Peg-QSVC, QNN, VQC) and feature mapping techniques (ZFeatureMap, ZZFeatureMap, and PauliFeatureMap). Results indicate that QSVC is notably robust under noise, whereas Peg-QSVC and QNN are more sensitive, particularly to depolarizing and amplitude-damping noise. The PauliFeatureMap is especially vulnerable, highlighting difficulties in maintaining accurate classification under noisy conditions. These findings underscore the critical importance of feature map selection and noise mitigation strategies in optimizing QML for genomic classification, with promising implications for personalized medicine.
☆ Data-driven system identification using quadratic embeddings of nonlinear dynamics
We propose a novel data-driven method called QENDy (Quadratic Embedding of Nonlinear Dynamics) that not only allows us to learn quadratic representations of highly nonlinear dynamical systems, but also to identify the governing equations. The approach is based on an embedding of the system into a higher-dimensional feature space in which the dynamics become quadratic. Just like SINDy (Sparse Identification of Nonlinear Dynamics), our method requires trajectory data, time derivatives for the training data points, which can also be estimated using finite difference approximations, and a set of preselected basis functions, called dictionary. We illustrate the efficacy and accuracy of QENDy with the aid of various benchmark problems and compare its performance with SINDy and a deep learning method for identifying quadratic embeddings. Furthermore, we analyze the convergence of QENDy and SINDy in the infinite data limit, highlight their similarities and main differences, and compare the quadratic embedding with linearization techniques based on the Koopman operator.
☆ Globally Convergent Variational Inference NeurIPS 2024
In variational inference (VI), an approximation of the posterior distribution is selected from a family of distributions through numerical optimization. With the most common variational objective function, known as the evidence lower bound (ELBO), only convergence to a local optimum can be guaranteed. In this work, we instead establish the global convergence of a particular VI method. This VI method, which may be considered an instance of neural posterior estimation (NPE), minimizes an expectation of the inclusive (forward) KL divergence to fit a variational distribution that is parameterized by a neural network. Our convergence result relies on the neural tangent kernel (NTK) to characterize the gradient dynamics that arise from considering the variational objective in function space. In the asymptotic regime of a fixed, positive-definite neural tangent kernel, we establish conditions under which the variational objective admits a unique solution in a reproducing kernel Hilbert space (RKHS). Then, we show that the gradient descent dynamics in function space converge to this unique function. In ablation studies and practical problems, we demonstrate that our results explain the behavior of NPE in non-asymptotic finite-neuron settings, and show that NPE outperforms ELBO-based optimization, which often converges to shallow local optima.
comment: Accepted to the 38th Conference on Neural Information Processing Systems (NeurIPS 2024)
☆ CWEval: Outcome-driven Evaluation on Functionality and Security of LLM Code Generation
Large Language Models (LLMs) have significantly aided developers by generating or assisting in code writing, enhancing productivity across various tasks. While identifying incorrect code is often straightforward, detecting vulnerabilities in functionally correct code is more challenging, especially for developers with limited security knowledge, which poses considerable security risks of using LLM-generated code and underscores the need for robust evaluation benchmarks that assess both functional correctness and security. Current benchmarks like CyberSecEval and SecurityEval attempt to solve it but are hindered by unclear and impractical specifications, failing to assess both functionality and security accurately. To tackle these deficiencies, we introduce CWEval, a novel outcome-driven evaluation framework designed to enhance the evaluation of secure code generation by LLMs. This framework not only assesses code functionality but also its security simultaneously with high-quality task specifications and outcome-driven test oracles which provides high accuracy. Coupled with CWEval-bench, a multilingual, security-critical coding benchmark, CWEval provides a rigorous empirical security evaluation on LLM-generated code, overcoming previous benchmarks' shortcomings. Through our evaluations, CWEval reveals a notable portion of functional but insecure code produced by LLMs, and shows a serious inaccuracy of previous evaluations, ultimately contributing significantly to the field of secure code generation. We open-source our artifact at: https://github.com/Co1lin/CWEval .
comment: to be published in LLM4Code 2025
Self-supervised Deep Hyperspectral Inpainting with the Plug and Play and Deep Image Prior Models
Hyperspectral images are typically composed of hundreds of narrow and contiguous spectral bands, each containing information regarding the material composition of the imaged scene. However, these images can be affected by various sources of noise, distortions, or data loss, which can significantly degrade their quality and usefulness. This paper introduces a convergent guaranteed algorithm, LRS-PnP-DIP(1-Lip), which successfully addresses the instability issue of DHP that has been reported before. The proposed algorithm extends the successful joint low-rank and sparse model to further exploit the underlying data structures beyond the conventional and sometimes restrictive unions of subspace models. A stability analysis guarantees the convergence of the proposed algorithm under mild assumptions , which is crucial for its application in real-world scenarios. Extensive experiments demonstrate that the proposed solution consistently delivers visually and quantitatively superior inpainting results, establishing state-of-the-art performance.
comment: 31 pages, 9 Figures, 7 Tables. arXiv admin note: text overlap with arXiv:2306.08128
☆ Modeling Quantum Machine Learning for Genomic Data Analysis
Quantum Machine Learning (QML) continues to evolve, unlocking new opportunities for diverse applications. In this study, we investigate and evaluate the applicability of QML models for binary classification of genome sequence data by employing various feature mapping techniques. We present an open-source, independent Qiskit-based implementation to conduct experiments on a benchmark genomic dataset. Our simulations reveal that the interplay between feature mapping techniques and QML algorithms significantly influences performance. Notably, the Pegasos Quantum Support Vector Classifier (Pegasos-QSVC) exhibits high sensitivity, particularly excelling in recall metrics, while Quantum Neural Networks (QNN) achieve the highest training accuracy across all feature maps. However, the pronounced variability in classifier performance, dependent on feature mapping, highlights the risk of overfitting to localized output distributions in certain scenarios. This work underscores the transformative potential of QML for genomic data classification while emphasizing the need for continued advancements to enhance the robustness and accuracy of these methodologies.
☆ A Critical Synthesis of Uncertainty Quantification and Foundation Models in Monocular Depth Estimation
While recent foundation models have enabled significant breakthroughs in monocular depth estimation, a clear path towards safe and reliable deployment in the real-world remains elusive. Metric depth estimation, which involves predicting absolute distances, poses particular challenges, as even the most advanced foundation models remain prone to critical errors. Since quantifying the uncertainty has emerged as a promising endeavor to address these limitations and enable trustworthy deployment, we fuse five different uncertainty quantification methods with the current state-of-the-art DepthAnythingV2 foundation model. To cover a wide range of metric depth domains, we evaluate their performance on four diverse datasets. Our findings identify fine-tuning with the Gaussian Negative Log-Likelihood Loss (GNLL) as a particularly promising approach, offering reliable uncertainty estimates while maintaining predictive performance and computational efficiency on par with the baseline, encompassing both training and inference time. By fusing uncertainty quantification and foundation models within the context of monocular depth estimation, this paper lays a critical foundation for future research aimed at improving not only model performance but also its explainability. Extending this critical synthesis of uncertainty quantification and foundation models into other crucial tasks, such as semantic segmentation and pose estimation, presents exciting opportunities for safer and more reliable machine vision systems.
☆ A Multi-Modal AI Copilot for Single-Cell Analysis with Instruction Following
Large language models excel at interpreting complex natural language instructions, enabling them to perform a wide range of tasks. In the life sciences, single-cell RNA sequencing (scRNA-seq) data serves as the "language of cellular biology", capturing intricate gene expression patterns at the single-cell level. However, interacting with this "language" through conventional tools is often inefficient and unintuitive, posing challenges for researchers. To address these limitations, we present InstructCell, a multi-modal AI copilot that leverages natural language as a medium for more direct and flexible single-cell analysis. We construct a comprehensive multi-modal instruction dataset that pairs text-based instructions with scRNA-seq profiles from diverse tissues and species. Building on this, we develop a multi-modal cell language architecture capable of simultaneously interpreting and processing both modalities. InstructCell empowers researchers to accomplish critical tasks-such as cell type annotation, conditional pseudo-cell generation, and drug sensitivity prediction-using straightforward natural language commands. Extensive evaluations demonstrate that InstructCell consistently meets or exceeds the performance of existing single-cell foundation models, while adapting to diverse experimental conditions. More importantly, InstructCell provides an accessible and intuitive tool for exploring complex single-cell data, lowering technical barriers and enabling deeper biological insights.
comment: 37 pages; 13 figures; Code: https://github.com/zjunlp/Instructcell; Models: https://huggingface.co/zjunlp/Instructcell-chat, https://huggingface.co/zjunlp/InstructCell-instruct
☆ D$^2$-DPM: Dual Denoising for Quantized Diffusion Probabilistic Models AAAI2025
Diffusion models have achieved cutting-edge performance in image generation. However, their lengthy denoising process and computationally intensive score estimation network impede their scalability in low-latency and resource-constrained scenarios. Post-training quantization (PTQ) compresses and accelerates diffusion models without retraining, but it inevitably introduces additional quantization noise, resulting in mean and variance deviations. In this work, we propose D2-DPM, a dual denoising mechanism aimed at precisely mitigating the adverse effects of quantization noise on the noise estimation network. Specifically, we first unravel the impact of quantization noise on the sampling equation into two components: the mean deviation and the variance deviation. The mean deviation alters the drift coefficient of the sampling equation, influencing the trajectory trend, while the variance deviation magnifies the diffusion coefficient, impacting the convergence of the sampling trajectory. The proposed D2-DPM is thus devised to denoise the quantization noise at each time step, and then denoise the noisy sample through the inverse diffusion iterations. Experimental results demonstrate that D2-DPM achieves superior generation quality, yielding a 1.42 lower FID than the full-precision model while achieving 3.99x compression and 11.67x bit-operation acceleration.
comment: 9 pages, 4 figures, acceptted by AAAI2025
☆ Revolutionizing Communication with Deep Learning and XAI for Enhanced Arabic Sign Language Recognition
This study introduces an integrated approach to recognizing Arabic Sign Language (ArSL) using state-of-the-art deep learning models such as MobileNetV3, ResNet50, and EfficientNet-B2. These models are further enhanced by explainable AI (XAI) techniques to boost interpretability. The ArSL2018 and RGB Arabic Alphabets Sign Language (AASL) datasets are employed, with EfficientNet-B2 achieving peak accuracies of 99.48\% and 98.99\%, respectively. Key innovations include sophisticated data augmentation methods to mitigate class imbalance, implementation of stratified 5-fold cross-validation for better generalization, and the use of Grad-CAM for clear model decision transparency. The proposed system not only sets new benchmarks in recognition accuracy but also emphasizes interpretability, making it suitable for applications in healthcare, education, and inclusive communication technologies.
comment: 13 pages, 25 figures, 16 tables
☆ Inference-Time-Compute: More Faithful? A Research Note
Models trained specifically to generate long Chains of Thought (CoTs) have recently achieved impressive results. We refer to these models as Inference-Time-Compute (ITC) models. Are the CoTs of ITC models more faithful compared to traditional non-ITC models? We evaluate two ITC models (based on Qwen-2.5 and Gemini-2) on an existing test of faithful CoT To measure faithfulness, we test if models articulate cues in their prompt that influence their answers to MMLU questions. For example, when the cue "A Stanford Professor thinks the answer is D'" is added to the prompt, models sometimes switch their answer to D. In such cases, the Gemini ITC model articulates the cue 54% of the time, compared to 14% for the non-ITC Gemini. We evaluate 7 types of cue, such as misleading few-shot examples and anchoring on past responses. ITC models articulate cues that influence them much more reliably than all the 6 non-ITC models tested, such as Claude-3.5-Sonnet and GPT-4o, which often articulate close to 0% of the time. However, our study has important limitations. We evaluate only two ITC models -- we cannot evaluate OpenAI's SOTA o1 model. We also lack details about the training of these ITC models, making it hard to attribute our findings to specific processes. We think faithfulness of CoT is an important property for AI Safety. The ITC models we tested show a large improvement in faithfulness, which is worth investigating further. To speed up this investigation, we release these early results as a research note.
comment: 7 pages, 5 figures
☆ FairTTTS: A Tree Test Time Simulation Method for Fairness-Aware Classification
Algorithmic decision-making has become deeply ingrained in many domains, yet biases in machine learning models can still produce discriminatory outcomes, often harming unprivileged groups. Achieving fair classification is inherently challenging, requiring a careful balance between predictive performance and ethical considerations. We present FairTTTS, a novel post-processing bias mitigation method inspired by the Tree Test Time Simulation (TTTS) method. Originally developed to enhance accuracy and robustness against adversarial inputs through probabilistic decision-path adjustments, TTTS serves as the foundation for FairTTTS. By building on this accuracy-enhancing technique, FairTTTS mitigates bias and improves predictive performance. FairTTTS uses a distance-based heuristic to adjust decisions at protected attribute nodes, ensuring fairness for unprivileged samples. This fairness-oriented adjustment occurs as a post-processing step, allowing FairTTTS to be applied to pre-trained models, diverse datasets, and various fairness metrics without retraining. Extensive evaluation on seven benchmark datasets shows that FairTTTS outperforms traditional methods in fairness improvement, achieving a 20.96% average increase over the baseline compared to 18.78% for related work, and further enhances accuracy by 0.55%. In contrast, competing methods typically reduce accuracy by 0.42%. These results confirm that FairTTTS effectively promotes more equitable decision-making while simultaneously improving predictive performance.
☆ Multiple-Input Variational Auto-Encoder for Anomaly Detection in Heterogeneous Data
Anomaly detection (AD) plays a pivotal role in AI applications, e.g., in classification, and intrusion/threat detection in cybersecurity. However, most existing methods face challenges of heterogeneity amongst feature subsets posed by non-independent and identically distributed (non-IID) data. We propose a novel neural network model called Multiple-Input Auto-Encoder for AD (MIAEAD) to address this. MIAEAD assigns an anomaly score to each feature subset of a data sample to indicate its likelihood of being an anomaly. This is done by using the reconstruction error of its sub-encoder as the anomaly score. All sub-encoders are then simultaneously trained using unsupervised learning to determine the anomaly scores of feature subsets. The final AUC of MIAEAD is calculated for each sub-dataset, and the maximum AUC obtained among the sub-datasets is selected. To leverage the modelling of the distribution of normal data to identify anomalies of the generative models, we develop a novel neural network architecture/model called Multiple-Input Variational Auto-Encoder (MIVAE). MIVAE can process feature subsets through its sub-encoders before learning distribution of normal data in the latent space. This allows MIVAE to identify anomalies that deviate from the learned distribution. We theoretically prove that the difference in the average anomaly score between normal samples and anomalies obtained by the proposed MIVAE is greater than that of the Variational Auto-Encoder (VAEAD), resulting in a higher AUC for MIVAE. Extensive experiments on eight real-world anomaly datasets demonstrate the superior performance of MIAEAD and MIVAE over conventional methods and the state-of-the-art unsupervised models, by up to 6% in terms of AUC score. Alternatively, MIAEAD and MIVAE have a high AUC when applied to feature subsets with low heterogeneity based on the coefficient of variation (CV) score.
comment: 16 pages
☆ Bootstrapping Corner Cases: High-Resolution Inpainting for Safety Critical Detect and Avoid for Automated Flying
Modern machine learning techniques have shown tremendous potential, especially for object detection on camera images. For this reason, they are also used to enable safety-critical automated processes such as autonomous drone flights. We present a study on object detection for Detect and Avoid, a safety critical function for drones that detects air traffic during automated flights for safety reasons. An ill-posed problem is the generation of good and especially large data sets, since detection itself is the corner case. Most models suffer from limited ground truth in raw data, \eg recorded air traffic or frontal flight with a small aircraft. It often leads to poor and critical detection rates. We overcome this problem by using inpainting methods to bootstrap the dataset such that it explicitly contains the corner cases of the raw data. We provide an overview of inpainting methods and generative models and present an example pipeline given a small annotated dataset. We validate our method by generating a high-resolution dataset, which we make publicly available and present it to an independent object detector that was fully trained on real data.
☆ EEG-ReMinD: Enhancing Neurodegenerative EEG Decoding through Self-Supervised State Reconstruction-Primed Riemannian Dynamics
The development of EEG decoding algorithms confronts challenges such as data sparsity, subject variability, and the need for precise annotations, all of which are vital for advancing brain-computer interfaces and enhancing the diagnosis of diseases. To address these issues, we propose a novel two-stage approach named Self-Supervised State Reconstruction-Primed Riemannian Dynamics (EEG-ReMinD) , which mitigates reliance on supervised learning and integrates inherent geometric features. This approach efficiently handles EEG data corruptions and reduces the dependency on labels. EEG-ReMinD utilizes self-supervised and geometric learning techniques, along with an attention mechanism, to analyze the temporal dynamics of EEG features within the framework of Riemannian geometry, referred to as Riemannian dynamics. Comparative analyses on both intact and corrupted datasets from two different neurodegenerative disorders underscore the enhanced performance of EEG-ReMinD.
☆ An Empirical Wall-Pressure Spectrum Model for Aeroacoustic Predictions Based on Symbolic Regression
Fast-turn around methods to predict airfoil trailing-edge noise are crucial for incorporating noise limitations into design optimization loops of several applications. Among these aeroacoustic predictive models, Amiet's theory offers the best balance between accuracy and simplicity. The accuracy of the model relies heavily on precise wall-pressure spectrum predictions, which are often based on single-equation formulations with adjustable parameters. These parameters are calibrated for particular airfoils and flow conditions and consequently tend to fail when applied outside their calibration range. This paper introduces a new wall-pressure spectrum empirical model designed to enhance the robustness and accuracy of current state-of-the-art predictions while widening the range of applicability of the model to different airfoils and flow conditions. The model is developed using AI-based symbolic regression via a genetic-algorithm-based approach, and applied to a dataset of wall-pressure fluctuations measured on NACA 0008 and NACA 63018 airfoils at multiple angles of attack and inflow velocities, covering turbulent boundary layers with both adverse and favorable pressure gradients. Validation against experimental data (outside the training dataset) demonstrates the robustness of the model compared to well-accepted semi-empirical models. Finally, the model is integrated with Amiet's theory to predict the aeroacoustic noise of a full-scale wind turbine, showing good agreement with experimental measurements.
☆ RoHan: Robust Hand Detection in Operation Room
Hand-specific localization has garnered significant interest within the computer vision community. Although there are numerous datasets with hand annotations from various angles and settings, domain transfer techniques frequently struggle in surgical environments. This is mainly due to the limited availability of gloved hand instances and the unique challenges of operating rooms (ORs). Thus, hand-detection models tailored to OR settings require extensive training and expensive annotation processes. To overcome these challenges, we present "RoHan" - a novel approach for robust hand detection in the OR, leveraging advanced semi-supervised domain adaptation techniques to tackle the challenges of varying recording conditions, diverse glove colors, and occlusions common in surgical settings. Our methodology encompasses two main stages: (1) data augmentation strategy that utilizes "Artificial Gloves," a method for augmenting publicly available hand datasets with synthetic images of hands-wearing gloves; (2) semi-supervised domain adaptation pipeline that improves detection performance in real-world OR settings through iterative prediction refinement and efficient frame filtering. We evaluate our method using two datasets: simulated enterotomy repair and saphenous vein graft harvesting. "RoHan" substantially reduces the need for extensive labeling and model training, paving the way for the practical implementation of hand detection technologies in medical settings.
comment: 12 pages
☆ Data-driven inventory management for new products: A warm-start and adjusted Dyna-$Q$ approach
In this paper, we propose a novel reinforcement learning algorithm for inventory management of newly launched products with no or limited historical demand information. The algorithm follows the classic Dyna-$Q$ structure, balancing the model-based and model-free approaches, while accelerating the training process of Dyna-$Q$ and mitigating the model discrepancy generated by the model-based feedback. Warm-start information from the demand data of existing similar products can be incorporated into the algorithm to further stabilize the early-stage training and reduce the variance of the estimated optimal policy. Our approach is validated through a case study of bakery inventory management with real data. The adjusted Dyna-$Q$ shows up to a 23.7\% reduction in average daily cost compared with $Q$-learning, and up to a 77.5\% reduction in training time within the same horizon compared with classic Dyna-$Q$. By incorporating the warm-start information, it can be found that the adjusted Dyna-$Q$ has the lowest total cost, lowest variance in total cost, and relatively low shortage percentages among all the algorithms under a 30-day testing.
comment: 7 pages, 2 figures
☆ Smooth Handovers via Smoothed Online Learning
With users demanding seamless connectivity, handovers (HOs) have become a fundamental element of cellular networks. However, optimizing HOs is a challenging problem, further exacerbated by the growing complexity of mobile networks. This paper presents the first countrywide study of HO optimization, through the prism of Smoothed Online Learning (SOL). We first analyze an extensive dataset from a commercial mobile network operator (MNO) in Europe with more than 40M users, to understand and reveal important features and performance impacts on HOs. Our findings highlight a correlation between HO failures/delays, and the characteristics of radio cells and end-user devices, showcasing the impact of heterogeneity in mobile networks nowadays. We subsequently model UE-cell associations as dynamic decisions and propose a realistic system model for smooth and accurate HOs that extends existing approaches by (i) incorporating device and cell features on HO optimization, and (ii) eliminating (prior) strong assumptions about requiring future signal measurements and knowledge of end-user mobility. Our algorithm, aligned with the O-RAN paradigm, provides robust dynamic regret guarantees, even in challenging environments, and shows superior performance in multiple scenarios with real-world and synthetic data.
☆ Hybrid Action Based Reinforcement Learning for Multi-Objective Compatible Autonomous Driving
Reinforcement Learning (RL) has shown excellent performance in solving decision-making and control problems of autonomous driving, which is increasingly applied in diverse driving scenarios. However, driving is a multi-attribute problem, leading to challenges in achieving multi-objective compatibility for current RL methods, especially in both policy execution and policy iteration. On the one hand, the common action space structure with single action type limits driving flexibility or results in large behavior fluctuations during policy execution. On the other hand, the multi-attribute weighted single reward function result in the agent's disproportionate attention to certain objectives during policy iterations. To this end, we propose a Multi-objective Ensemble-Critic reinforcement learning method with Hybrid Parametrized Action for multi-objective compatible autonomous driving. Specifically, a parameterized action space is constructed to generate hybrid driving actions, combining both abstract guidance and concrete control commands. A multi-objective critics architecture is constructed considering multiple attribute rewards, to ensure simultaneously focusing on different driving objectives. Additionally, uncertainty-based exploration strategy is introduced to help the agent faster approach viable driving policy. The experimental results in both the simulated traffic environment and the HighD dataset demonstrate that our method can achieve multi-objective compatible autonomous driving in terms of driving efficiency, action consistency, and safety. It enhances the general performance of the driving while significantly increasing training efficiency.
comment: 12 pages, 9 figures, 5 tables
☆ Dynamic Multimodal Sentiment Analysis: Leveraging Cross-Modal Attention for Enabled Classification
This paper explores the development of a multimodal sentiment analysis model that integrates text, audio, and visual data to enhance sentiment classification. The goal is to improve emotion detection by capturing the complex interactions between these modalities, thereby enabling more accurate and nuanced sentiment interpretation. The study evaluates three feature fusion strategies -- late stage fusion, early stage fusion, and multi-headed attention -- within a transformer-based architecture. Experiments were conducted using the CMU-MOSEI dataset, which includes synchronized text, audio, and visual inputs labeled with sentiment scores. Results show that early stage fusion significantly outperforms late stage fusion, achieving an accuracy of 71.87\%, while the multi-headed attention approach offers marginal improvement, reaching 72.39\%. The findings suggest that integrating modalities early in the process enhances sentiment classification, while attention mechanisms may have limited impact within the current framework. Future work will focus on refining feature fusion techniques, incorporating temporal data, and exploring dynamic feature weighting to further improve model performance.
☆ CuAsmRL: Optimizing GPU SASS Schedules via Deep Reinforcement Learning
Large language models (LLMs) are remarked by their substantial computational requirements. To mitigate the cost, researchers develop specialized CUDA kernels, which often fuse several tensor operations to maximize the utilization of GPUs as much as possible. However, those specialized kernels may still leave performance on the table as CUDA assembly experts show that manual optimization of GPU SASS schedules can lead to better performance, and trial-and-error is largely employed to manually find the best GPU SASS schedules. In this work, we employ an automatic approach to optimize GPU SASS schedules, which thus can be integrated into existing compiler frameworks. The key to automatic optimization is training an RL agent to mimic how human experts perform manual scheduling. To this end, we formulate an assembly game, where RL agents can play to find the best GPU SASS schedules. The assembly game starts from a \textit{-O3} optimized SASS schedule, and the RL agents can iteratively apply actions to mutate the current schedules. Positive rewards are generated if the mutated schedules get higher throughput by executing on GPUs. Experiments show that CuAsmRL can further improve the performance of existing specialized CUDA kernels transparently by up to $26\%$, and on average $9\%$. Moreover, it is used as a tool to reveal potential optimization moves learned automatically.
comment: cgo 2025
☆ Optimal Policy Adaptation under Covariate Shift
Transfer learning of prediction models has been extensively studied, while the corresponding policy learning approaches are rarely discussed. In this paper, we propose principled approaches for learning the optimal policy in the target domain by leveraging two datasets: one with full information from the source domain and the other from the target domain with only covariates. First, under the setting of covariate shift, we formulate the problem from a perspective of causality and present the identifiability assumptions for the reward induced by a given policy. Then, we derive the efficient influence function and the semiparametric efficiency bound for the reward. Based on this, we construct a doubly robust and semiparametric efficient estimator for the reward and then learn the optimal policy by optimizing the estimated reward. Moreover, we theoretically analyze the bias and the generalization error bound for the learned policy. Furthermore, in the presence of both covariate and concept shifts, we propose a novel sensitivity analysis method to evaluate the robustness of the proposed policy learning approach. Extensive experiments demonstrate that the approach not only estimates the reward more accurately but also yields a policy that closely approximates the theoretically optimal policy.
☆ On the use of Statistical Learning Theory for model selection in Structural Health Monitoring
Whenever data-based systems are employed in engineering applications, defining an optimal statistical representation is subject to the problem of model selection. This paper focusses on how well models can generalise in Structural Health Monitoring (SHM). Although statistical model validation in this field is often performed heuristically, it is possible to estimate generalisation more rigorously using the bounds provided by Statistical Learning Theory (SLT). Therefore, this paper explores the selection process of a kernel smoother for modelling the impulse response of a linear oscillator from the perspective of SLT. It is demonstrated that incorporating domain knowledge into the regression problem yields a lower guaranteed risk, thereby enhancing generalisation.
☆ Self-Attentive Spatio-Temporal Calibration for Precise Intermediate Layer Matching in ANN-to-SNN Distillation
Spiking Neural Networks (SNNs) are promising for low-power computation due to their event-driven mechanism but often suffer from lower accuracy compared to Artificial Neural Networks (ANNs). ANN-to-SNN knowledge distillation can improve SNN performance, but previous methods either focus solely on label information, missing valuable intermediate layer features, or use a layer-wise approach that neglects spatial and temporal semantic inconsistencies, leading to performance degradation.To address these limitations, we propose a novel method called self-attentive spatio-temporal calibration (SASTC). SASTC uses self-attention to identify semantically aligned layer pairs between ANN and SNN, both spatially and temporally. This enables the autonomous transfer of relevant semantic information. Extensive experiments show that SASTC outperforms existing methods, effectively solving the mismatching problem. Superior accuracy results include 95.12% on CIFAR-10, 79.40% on CIFAR-100 with 2 time steps, and 68.69% on ImageNet with 4 time steps for static datasets, and 97.92% on DVS-Gesture and 83.60% on DVS-CIFAR10 for neuromorphic datasets. This marks the first time SNNs have outperformed ANNs on both CIFAR-10 and CIFAR-100, shedding the new light on the potential applications of SNNs.
☆ Gen-A: Generalizing Ambisonics Neural Encoding to Unseen Microphone Arrays
Using deep neural networks (DNNs) for encoding of microphone array (MA) signals to the Ambisonics spatial audio format can surpass certain limitations of established conventional methods, but existing DNN-based methods need to be trained separately for each MA. This paper proposes a DNN-based method for Ambisonics encoding that can generalize to arbitrary MA geometries unseen during training. The method takes as inputs the MA geometry and MA signals and uses a multi-level encoder consisting of separate paths for geometry and signal data, where geometry features inform the signal encoder at each level. The method is validated in simulated anechoic and reverberant conditions with one and two sources. The results indicate improvement over conventional encoding across the whole frequency range for dry scenes, while for reverberant scenes the improvement is frequency-dependent.
comment: Accepted for publication in Proceedings of the 2025 IEEE International Conference on Acoustics, Speech and Signal Processing
☆ UFGraphFR: An attempt at a federated recommendation system based on user text characteristics
Federated learning has become an important research area in 'private computing' due to the 'useable invisibility' of data during training. Inspired by Federated learning, the federated recommendation system has gradually become a new recommendation service architecture that can protect users' privacy. The use of user diagrams to enhance federated recommendations is a promising topic. How to use user diagrams to enhance federated recommendations is a promising research topic. However, it's a great challenge to construct a user diagram without compromising privacy in a federated learning scenario. Inspired by the simple idea that similar users often have the same attribute characteristics, we propose a personalized federated recommendation algorithm based on the user relationship graph constructed by the user text characteristics(Graph Federation Recommendation System based on User Text description Features, UFGraphFR). The method uses the embedding layer weight of the user's text feature description to construct the user relationship graph. It introduces the Transformer mechanism to capture the sequence modeling of the user's historical interaction sequence. Without access to user history interactions and specific user attributes, the federal learning privacy protection of data 'useable invisibility' is embodied. Preliminary experiments on some benchmark datasets demonstrate the superior performance of UFGraphFR. Our experiments show that this model can protect user privacy to some extent without affecting the performance of the recommendation system. The code will be easily available on https://github.com/trueWangSyutung/UFGraphFR.
☆ PolyLUT: Ultra-low Latency Polynomial Inference with Hardware-Aware Structured Pruning
Standard deep neural network inference involves the computation of interleaved linear maps and nonlinear activation functions. Prior work for ultra-low latency implementations has hardcoded these operations inside FPGA lookup tables (LUTs). However, FPGA LUTs can implement a much greater variety of functions. In this paper, we propose a novel approach to training DNNs for FPGA deployment using multivariate polynomials as the basic building block. Our method takes advantage of the flexibility offered by the soft logic, hiding the polynomial evaluation inside the LUTs with minimal overhead. By using polynomial building blocks, we achieve the same accuracy using considerably fewer layers of soft logic than by using linear functions, leading to significant latency and area improvements. LUT-based implementations also face a significant challenge: the LUT size grows exponentially with the number of inputs. Prior work relies on a priori fixed sparsity, with results heavily dependent on seed selection. To address this, we propose a structured pruning strategy using a bespoke hardware-aware group regularizer that encourages a particular sparsity pattern that leads to a small number of inputs per neuron. We demonstrate the effectiveness of PolyLUT on three tasks: network intrusion detection, jet identification at the CERN Large Hadron Collider, and MNIST.
comment: arXiv admin note: text overlap with arXiv:2309.02334
☆ Convergence Analysis of Real-time Recurrent Learning (RTRL) for a class of Recurrent Neural Networks
Recurrent neural networks (RNNs) are commonly trained with the truncated backpropagation-through-time (TBPTT) algorithm. For the purposes of computational tractability, the TBPTT algorithm truncates the chain rule and calculates the gradient on a finite block of the overall data sequence. Such approximation could lead to significant inaccuracies, as the block length for the truncated backpropagation is typically limited to be much smaller than the overall sequence length. In contrast, Real-time recurrent learning (RTRL) is an online optimization algorithm which asymptotically follows the true gradient of the loss on the data sequence as the number of sequence time steps $t \rightarrow \infty$. RTRL forward propagates the derivatives of the RNN hidden/memory units with respect to the parameters and, using the forward derivatives, performs online updates of the parameters at each time step in the data sequence. RTRL's online forward propagation allows for exact optimization over extremely long data sequences, although it can be computationally costly for models with large numbers of parameters. We prove convergence of the RTRL algorithm for a class of RNNs. The convergence analysis establishes a fixed point for the joint distribution of the data sequence, RNN hidden layer, and the RNN hidden layer forward derivatives as the number of data samples from the sequence and the number of training steps tend to infinity. We prove convergence of the RTRL algorithm to a stationary point of the loss. Numerical studies illustrate our theoretical results. One potential application area for RTRL is the analysis of financial data, which typically involve long time series and models with small to medium numbers of parameters. This makes RTRL computationally tractable and a potentially appealing optimization method for training models. Thus, we include an example of RTRL applied to limit order book data.
☆ Enhanced SPS Velocity-adaptive Scheme: Access Fariness in 5G NR V2I Networks SP
Vehicle-to-Infrastructure (V2I) technology enables information exchange between vehicles and road infrastructure. Specifically, when a vehicle approaches a roadside unit (RSU), it can exchange information with the RSU to obtain accurate data that assists in driving. With the release of the 3rd Generation Partnership Project (3GPP) Release 16, which includes the 5G New Radio (NR) Vehicle-to-Everything (V2X) standards, vehicles typically adopt mode-2 communication using sensing-based semi-persistent scheduling (SPS) for resource allocation. In this approach, vehicles identify candidate resources within a selection window and exclude ineligible resources based on information from a sensing window. However, vehicles often drive at different speeds, resulting in varying amounts of data transmission with RSUs as they pass by, which leads to unfair access. Therefore, it is essential to design an access scheme that accounts for different vehicle speeds to achieve fair access across the network. This paper formulates an optimization problem for vehicular networks and proposes a multi-objective optimization scheme to address it by adjusting the selection window in the SPS mechanism of 5G NR V2I mode-2. Simulation results demonstrate the effectiveness of the proposed scheme
comment: This paper has been submitted to IEEE Journal. The source code has been released at: https://github.com/qiongwu86/Enhanced-SPS-Velocity-adaptiveScheme-Access-Fariness-in-5G-NR-V2I-Networks
☆ An AI-driven framework for rapid and localized optimizations of urban open spaces
As urbanization accelerates, open spaces are increasingly recognized for their role in enhancing sustainability and well-being, yet they remain underexplored compared to built spaces. This study introduces an AI-driven framework that integrates machine learning models (MLMs) and explainable AI techniques to optimize Sky View Factor (SVF) and visibility, key spatial metrics influencing thermal comfort and perceived safety in urban spaces. Unlike global optimization methods, which are computationally intensive and impractical for localized adjustments, this framework supports incremental design improvements with lower computational costs and greater flexibility. The framework employs SHapley Adaptive Explanations (SHAP) to analyze feature importance and Counterfactual Explanations (CFXs) to propose minimal design changes. Simulations tested five MLMs, identifying XGBoost as the most accurate, with building width, park area, and heights of surrounding buildings as critical for SVF, and distances from southern buildings as key for visibility. Compared to Genetic Algorithms, which required approximately 15/30 minutes across 3/4 generations to converge, the tested CFX approach achieved optimized results in 1 minute with a 5% RMSE error, demonstrating significantly faster performance and suitability for scalable retrofitting strategies. This interpretable and computationally efficient framework advances urban performance optimization, providing data-driven insights and practical retrofitting solutions for enhancing usability and environmental quality across diverse urban contexts.
comment: 36 pages
☆ Maximizing Uncertainty for Federated learning via Bayesian Optimisation-based Model Poisoning
As we transition from Narrow Artificial Intelligence towards Artificial Super Intelligence, users are increasingly concerned about their privacy and the trustworthiness of machine learning (ML) technology. A common denominator for the metrics of trustworthiness is the quantification of uncertainty inherent in DL algorithms, and specifically in the model parameters, input data, and model predictions. One of the common approaches to address privacy-related issues in DL is to adopt distributed learning such as federated learning (FL), where private raw data is not shared among users. Despite the privacy-preserving mechanisms in FL, it still faces challenges in trustworthiness. Specifically, the malicious users, during training, can systematically create malicious model parameters to compromise the models predictive and generative capabilities, resulting in high uncertainty about their reliability. To demonstrate malicious behaviour, we propose a novel model poisoning attack method named Delphi which aims to maximise the uncertainty of the global model output. We achieve this by taking advantage of the relationship between the uncertainty and the model parameters of the first hidden layer of the local model. Delphi employs two types of optimisation , Bayesian Optimisation and Least Squares Trust Region, to search for the optimal poisoned model parameters, named as Delphi-BO and Delphi-LSTR. We quantify the uncertainty using the KL Divergence to minimise the distance of the predictive probability distribution towards an uncertain distribution of model output. Furthermore, we establish a mathematical proof for the attack effectiveness demonstrated in FL. Numerical results demonstrate that Delphi-BO induces a higher amount of uncertainty than Delphi-LSTR highlighting vulnerability of FL systems to model poisoning attacks.
comment: 14 pages
☆ Unsupervised Feature Construction for Anomaly Detection in Time Series -- An Evaluation
To detect anomalies with precision and without prior knowledge in time series, is it better to build a detector from the initial temporal representation, or to compute a new (tabular) representation using an existing automatic variable construction library? In this article, we address this question by conducting an in-depth experimental study for two popular detectors (Isolation Forest and Local Outlier Factor). The obtained results, for 5 different datasets, show that the new representation, computed using the tsfresh library, allows Isolation Forest to significantly improve its performance.
comment: 7
☆ Reward Compatibility: A Framework for Inverse RL
We provide an original theoretical study of Inverse Reinforcement Learning (IRL) through the lens of reward compatibility, a novel framework to quantify the compatibility of a reward with the given expert's demonstrations. Intuitively, a reward is more compatible with the demonstrations the closer the performance of the expert's policy computed with that reward is to the optimal performance for that reward. This generalizes the notion of feasible reward set, the most common framework in the theoretical IRL literature, for which a reward is either compatible or not compatible. The grayscale introduced by the reward compatibility is the key to extend the realm of provably efficient IRL far beyond what is attainable with the feasible reward set: from tabular to large-scale MDPs. We analyze the IRL problem across various settings, including optimal and suboptimal expert's demonstrations and both online and offline data collection. For all of these dimensions, we provide a tractable algorithm and corresponding sample complexity analysis, as well as various insights on reward compatibility and how the framework can pave the way to yet more general problem settings.
☆ Combining imaging and shape features for prediction tasks of Alzheimer's disease classification and brain age regression
We investigate combining imaging and shape features extracted from MRI for the clinically relevant tasks of brain age prediction and Alzheimer's disease classification. Our proposed model fuses ResNet-extracted image embeddings with shape embeddings from a bespoke graph neural network. The shape embeddings are derived from surface meshes of 15 brain structures, capturing detailed geometric information. Combined with the appearance features from T1-weighted images, we observe improvements in the prediction performance on both tasks, with substantial gains for classification. We evaluate the model using public datasets, including CamCAN, IXI, and OASIS3, demonstrating the effectiveness of fusing imaging and shape features for brain analysis.
☆ CHEQ-ing the Box: Safe Variable Impedance Learning for Robotic Polishing
Robotic systems are increasingly employed for industrial automation, with contact-rich tasks like polishing requiring dexterity and compliant behaviour. These tasks are difficult to model, making classical control challenging. Deep reinforcement learning (RL) offers a promising solution by enabling the learning of models and control policies directly from data. However, its application to real-world problems is limited by data inefficiency and unsafe exploration. Adaptive hybrid RL methods blend classical control and RL adaptively, combining the strengths of both: structure from control and learning from RL. This has led to improvements in data efficiency and exploration safety. However, their potential for hardware applications remains underexplored, with no evaluations on physical systems to date. Such evaluations are critical to fully assess the practicality and effectiveness of these methods in real-world settings. This work presents an experimental demonstration of the hybrid RL algorithm CHEQ for robotic polishing with variable impedance, a task requiring precise force and velocity tracking. In simulation, we show that variable impedance enhances polishing performance. We compare standalone RL with adaptive hybrid RL, demonstrating that CHEQ achieves effective learning while adhering to safety constraints. On hardware, CHEQ achieves effective polishing behaviour, requiring only eight hours of training and incurring just five failures. These results highlight the potential of adaptive hybrid RL for real-world, contact-rich tasks trained directly on hardware.
☆ Derivation of Output Correlation Inferences for Multi-Output (aka Multi-Task) Gaussian Process
Gaussian process (GP) is arguably one of the most widely used machine learning algorithms in practice. One of its prominent applications is Bayesian optimization (BO). Although the vanilla GP itself is already a powerful tool for BO, it is often beneficial to be able to consider the dependencies of multiple outputs. To do so, Multi-task GP (MTGP) is formulated, but it is not trivial to fully understand the derivations of its formulations and their gradients from the previous literature. This paper serves friendly derivations of the MTGP formulations and their gradients.
☆ AI Guide Dog: Egocentric Path Prediction on Smartphone
This paper introduces AI Guide Dog (AIGD), a lightweight egocentric navigation assistance system for visually impaired individuals, designed for real-time deployment on smartphones. AIGD addresses key challenges in blind navigation by employing a vision-only, multi-label classification approach to predict directional commands, ensuring safe traversal across diverse environments. We propose a novel technique to enable goal-based outdoor navigation by integrating GPS signals and high-level directions, while also addressing uncertain multi-path predictions for destination-free indoor navigation. Our generalized model is the first navigation assistance system to handle both goal-oriented and exploratory navigation scenarios across indoor and outdoor settings, establishing a new state-of-the-art in blind navigation. We present methods, datasets, evaluations, and deployment insights to encourage further innovations in assistive navigation systems.
☆ Gandalf the Red: Adaptive Security for LLMs
Current evaluations of defenses against prompt attacks in large language model (LLM) applications often overlook two critical factors: the dynamic nature of adversarial behavior and the usability penalties imposed on legitimate users by restrictive defenses. We propose D-SEC (Dynamic Security Utility Threat Model), which explicitly separates attackers from legitimate users, models multi-step interactions, and rigorously expresses the security-utility in an optimizable form. We further address the shortcomings in existing evaluations by introducing Gandalf, a crowd-sourced, gamified red-teaming platform designed to generate realistic, adaptive attack datasets. Using Gandalf, we collect and release a dataset of 279k prompt attacks. Complemented by benign user data, our analysis reveals the interplay between security and utility, showing that defenses integrated in the LLM (e.g., system prompts) can degrade usability even without blocking requests. We demonstrate that restricted application domains, defense-in-depth, and adaptive defenses are effective strategies for building secure and useful LLM applications. Code is available at \href{https://github.com/lakeraai/dsec-gandalf}{\texttt{https://github.com/lakeraai/dsec-gandalf}}.
comment: Niklas Pfister, V\'aclav Volhejn and Manuel Knott contributed equally
☆ Phase of Flight Classification in Aviation Safety using LSTM, GRU, and BiLSTM: A Case Study with ASN Dataset
Safety is the main concern in the aviation industry, where even minor operational issues can lead to serious consequences. This study addresses the need for comprehensive aviation accident analysis by leveraging natural language processing (NLP) and advanced AI models to classify the phase of flight from unstructured aviation accident analysis narratives. The research aims to determine whether the phase of flight can be inferred from narratives of post-accident events using NLP techniques. The classification performance of various deep learning models was evaluated. For single RNN-based models, LSTM achieved an accuracy of 63%, precision 60%, and recall 61%. BiLSTM recorded an accuracy of 64%, precision 63%, and a recall of 64%. GRU exhibited balanced performance with an accuracy and recall of 60% and a precision of 63%. Joint RNN-based models further enhanced predictive capabilities. GRU-LSTM, LSTM-BiLSTM, and GRU-BiLSTM demonstrated accuracy rates of 62%, 67%, and 60%, respectively, showcasing the benefits of combining these architectures. To provide a comprehensive overview of model performance, single and combined models were compared in terms of the various metrics. These results underscore the models' capacity to classify the phase of flight from raw text narratives, equipping aviation industry stakeholders with valuable insights for proactive decision-making. Therefore, this research signifies a substantial advancement in the application of NLP and deep learning models to enhance aviation safety.
comment: Aviation Safety, Deep learning algorithms, Flight phase, NLP, ASN, and Classification
☆ Aviation Safety Enhancement via NLP & Deep Learning: Classifying Flight Phases in ATSB Safety Reports
Aviation safety is paramount, demanding precise analysis of safety occurrences during different flight phases. This study employs Natural Language Processing (NLP) and Deep Learning models, including LSTM, CNN, Bidirectional LSTM (BLSTM), and simple Recurrent Neural Networks (sRNN), to classify flight phases in safety reports from the Australian Transport Safety Bureau (ATSB). The models exhibited high accuracy, precision, recall, and F1 scores, with LSTM achieving the highest performance of 87%, 88%, 87%, and 88%, respectively. This performance highlights their effectiveness in automating safety occurrence analysis. The integration of NLP and Deep Learning technologies promises transformative enhancements in aviation safety analysis, enabling targeted safety measures and streamlined report handling.
comment: NLP, Aviation Safety, ATSB, Deep learning, Flight phase. arXiv admin note: substantial text overlap with arXiv:2501.01694
☆ Logarithmic Memory Networks (LMNs): Efficient Long-Range Sequence Modeling for Resource-Constrained Environments
Long-range sequence modeling is a crucial aspect of natural language processing and time series analysis. However, traditional models like Recurrent Neural Networks (RNNs) and Transformers suffer from computational and memory inefficiencies, especially when dealing with long sequences. This paper introduces Logarithmic Memory Networks (LMNs), a novel architecture that leverages a hierarchical logarithmic tree structure to efficiently store and retrieve past information. LMNs dynamically summarize historical context, significantly reducing the memory footprint and computational complexity of attention mechanisms from O(n2) to O(log(n)). The model employs a single-vector, targeted attention mechanism to access stored information, and the memory block construction worker (summarizer) layer operates in two modes: a parallel execution mode during training for efficient processing of hierarchical tree structures and a sequential execution mode during inference, which acts as a memory management system. It also implicitly encodes positional information, eliminating the need for explicit positional encodings. These features make LMNs a robust and scalable solution for processing long-range sequences in resource-constrained environments, offering practical improvements in efficiency and scalability. The code is publicly available under the MIT License on GitHub: https://github.com/AhmedBoin/LogarithmicMemory.
comment: 18 pages, 10 figures
☆ Optimal Classification Trees for Continuous Feature Data Using Dynamic Programming with Branch-and-Bound AAAI-25
Computing an optimal classification tree that provably maximizes training performance within a given size limit, is NP-hard, and in practice, most state-of-the-art methods do not scale beyond computing optimal trees of depth three. Therefore, most methods rely on a coarse binarization of continuous features to maintain scalability. We propose a novel algorithm that optimizes trees directly on the continuous feature data using dynamic programming with branch-and-bound. We develop new pruning techniques that eliminate many sub-optimal splits in the search when similar to previously computed splits and we provide an efficient subroutine for computing optimal depth-two trees. Our experiments demonstrate that these techniques improve runtime by one or more orders of magnitude over state-of-the-art optimal methods and improve test accuracy by 5% over greedy heuristics.
comment: In the proceedings of AAAI-25
☆ Iterative Label Refinement Matters More than Preference Optimization under Weak Supervision
Language model (LM) post-training relies on two stages of human supervision: task demonstrations for supervised finetuning (SFT), followed by preference comparisons for reinforcement learning from human feedback (RLHF). As LMs become more capable, the tasks they are given become harder to supervise. Will post-training remain effective under unreliable supervision? To test this, we simulate unreliable demonstrations and comparison feedback using small LMs and time-constrained humans. We find that in the presence of unreliable supervision, SFT still retains some effectiveness, but DPO (a common RLHF algorithm) fails to improve the model beyond SFT. To address this, we propose iterative label refinement (ILR) as an alternative to RLHF. ILR improves the SFT data by using comparison feedback to decide whether human demonstrations should be replaced by model-generated alternatives, then retrains the model via SFT on the updated data. SFT+ILR outperforms SFT+DPO on several tasks with unreliable supervision (math, coding, and safe instruction-following). Our findings suggest that as LMs are used for complex tasks where human supervision is unreliable, RLHF may no longer be the best use of human comparison feedback; instead, it is better to direct feedback towards improving the training data rather than continually training the model. Our code and data are available at https://github.com/helloelwin/iterative-label-refinement.
comment: 22 pages, 10 figures
☆ Mitigating Algorithmic Bias in Multiclass CNN Classifications Using Causal Modeling
This study describes a procedure for applying causal modeling to detect and mitigate algorithmic bias in a multiclass classification problem. The dataset was derived from the FairFace dataset, supplemented with emotional labels generated by the DeepFace pre-trained model. A custom Convolutional Neural Network (CNN) was developed, consisting of four convolutional blocks, followed by fully connected layers and dropout layers to mitigate overfitting. Gender bias was identified in the CNN model's classifications: Females were more likely to be classified as "happy" or "sad," while males were more likely to be classified as "neutral." To address this, the one-vs-all (OvA) technique was applied. A causal model was constructed for each emotion class to adjust the CNN model's predicted class probabilities. The adjusted probabilities for the various classes were then aggregated by selecting the class with the highest probability. The resulting debiased classifications demonstrated enhanced gender fairness across all classes, with negligible impact--or even a slight improvement--on overall accuracy. This study highlights that algorithmic fairness and accuracy are not necessarily trade-offs. All data and code for this study are publicly available for download.
comment: 7 pages; 6 figures
☆ MD-Syn: Synergistic drug combination prediction based on the multidimensional feature fusion method and attention mechanisms
Drug combination therapies have shown promising therapeutic efficacy in complex diseases and have demonstrated the potential to reduce drug resistance. However, the huge number of possible drug combinations makes it difficult to screen them all in traditional experiments. In this study, we proposed MD-Syn, a computational framework, which is based on the multidimensional feature fusion method and multi-head attention mechanisms. Given drug pair-cell line triplets, MD-Syn considers one-dimensional and two-dimensional feature spaces simultaneously. It consists of a one-dimensional feature embedding module (1D-FEM), a two-dimensional feature embedding module (2D-FEM), and a deep neural network-based classifier for synergistic drug combination prediction. MD-Syn achieved the AUROC of 0.919 in 5-fold cross-validation, outperforming the state-of-the-art methods. Further, MD-Syn showed comparable results over two independent datasets. In addition, the multi-head attention mechanisms not only learn embeddings from different feature aspects but also focus on essential interactive feature elements, improving the interpretability of MD-Syn. In summary, MD-Syn is an interpretable framework to prioritize synergistic drug combination pairs with chemicals and cancer cell line gene expression profiles. To facilitate broader community access to this model, we have developed a web portal (https://labyeh104-2.life.nthu.edu.tw/) that enables customized predictions of drug combination synergy effects based on user-specified compounds.
☆ Distributed Nonparametric Estimation: from Sparse to Dense Samples per Terminal
Consider the communication-constrained problem of nonparametric function estimation, in which each distributed terminal holds multiple i.i.d. samples. Under certain regularity assumptions, we characterize the minimax optimal rates for all regimes, and identify phase transitions of the optimal rates as the samples per terminal vary from sparse to dense. This fully solves the problem left open by previous works, whose scopes are limited to regimes with either dense samples or a single sample per terminal. To achieve the optimal rates, we design a layered estimation protocol by exploiting protocols for the parametric density estimation problem. We show the optimality of the protocol using information-theoretic methods and strong data processing inequalities, and incorporating the classic balls and bins model. The optimal rates are immediate for various special cases such as density estimation, Gaussian, binary, Poisson and heteroskedastic regression models.
☆ deepTerra -- AI Land Classification Made Easy
deepTerra is a comprehensive platform designed to facilitate the classification of land surface features using machine learning and satellite imagery. The platform includes modules for data collection, image augmentation, training, testing, and prediction, streamlining the entire workflow for image classification tasks. This paper presents a detailed overview of the capabilities of deepTerra, shows how it has been applied to various research areas, and discusses the future directions it might take.
☆ State-of-the-Art Transformer Models for Image Super-Resolution: Techniques, Challenges, and Applications
Image Super-Resolution (SR) aims to recover a high-resolution image from its low-resolution counterpart, which has been affected by a specific degradation process. This is achieved by enhancing detail and visual quality. Recent advancements in transformer-based methods have remolded image super-resolution by enabling high-quality reconstructions surpassing previous deep-learning approaches like CNN and GAN-based. This effectively addresses the limitations of previous methods, such as limited receptive fields, poor global context capture, and challenges in high-frequency detail recovery. Additionally, the paper reviews recent trends and advancements in transformer-based SR models, exploring various innovative techniques and architectures that combine transformers with traditional networks to balance global and local contexts. These neoteric methods are critically analyzed, revealing promising yet unexplored gaps and potential directions for future research. Several visualizations of models and techniques are included to foster a holistic understanding of recent trends. This work seeks to offer a structured roadmap for researchers at the forefront of deep learning, specifically exploring the impact of transformers on super-resolution techniques.
comment: 8 pages
☆ An Intra- and Cross-frame Topological Consistency Scheme for Semi-supervised Atherosclerotic Coronary Plaque Segmentation ICASSP 2025
Enhancing the precision of segmenting coronary atherosclerotic plaques from CT Angiography (CTA) images is pivotal for advanced Coronary Atherosclerosis Analysis (CAA), which distinctively relies on the analysis of vessel cross-section images reconstructed via Curved Planar Reformation. This task presents significant challenges due to the indistinct boundaries and structures of plaques and blood vessels, leading to the inadequate performance of current deep learning models, compounded by the inherent difficulty in annotating such complex data. To address these issues, we propose a novel dual-consistency semi-supervised framework that integrates Intra-frame Topological Consistency (ITC) and Cross-frame Topological Consistency (CTC) to leverage labeled and unlabeled data. ITC employs a dual-task network for simultaneous segmentation mask and Skeleton-aware Distance Transform (SDT) prediction, achieving similar prediction of topology structure through consistency constraint without additional annotations. Meanwhile, CTC utilizes an unsupervised estimator for analyzing pixel flow between skeletons and boundaries of adjacent frames, ensuring spatial continuity. Experiments on two CTA datasets show that our method surpasses existing semi-supervised methods and approaches the performance of supervised methods on CAA. In addition, our method also performs better than other methods on the ACDC dataset, demonstrating its generalization.
comment: Accepted by ICASSP 2025
☆ Flow: A Modular Approach to Automated Agentic Workflow Generation
Multi-agent frameworks powered by large language models (LLMs) have demonstrated great success in automated planning and task execution. However, the effective adjustment of Agentic workflows during execution has not been well-studied. A effective workflow adjustment is crucial, as in many real-world scenarios, the initial plan must adjust to unforeseen challenges and changing conditions in real-time to ensure the efficient execution of complex tasks. In this paper, we define workflows as an activity-on-vertex (AOV) graphs. We continuously refine the workflow by dynamically adjusting task allocations based on historical performance and previous AOV with LLM agents. To further enhance system performance, we emphasize modularity in workflow design based on measuring parallelism and dependence complexity. Our proposed multi-agent framework achieved efficient sub-task concurrent execution, goal achievement, and error tolerance. Empirical results across different practical tasks demonstrate dramatic improvements in the efficiency of multi-agent frameworks through dynamic workflow updating and modularization.
☆ Prediction Interval Construction Method for Electricity Prices
Accurate prediction of electricity prices plays an essential role in the electricity market. To reflect the uncertainty of electricity prices, price intervals are predicted. This paper proposes a novel prediction interval construction method. A conditional generative adversarial network is first presented to generate electricity price scenarios, with which the prediction intervals can be constructed. Then, different generated scenarios are stacked to obtain the probability densities, which can be applied to accurately reflect the uncertainty of electricity prices. Furthermore, a reinforced prediction mechanism based on the volatility level of weather factors is introduced to address the spikes or volatile prices. A case study is conducted to verify the effectiveness of the proposed novel prediction interval construction method. The method can also provide the probability density of each price scenario within the prediction interval and has the superiority to address the volatile prices and price spikes with a reinforced prediction mechanism.
☆ Real-time Verification and Refinement of Language Model Text Generation
Large language models (LLMs) have shown remarkable performance across a wide range of natural language tasks. However, a critical challenge remains in that they sometimes generate factually incorrect answers. To address this, while many previous work has focused on identifying errors in their generation and further refining them, they are slow in deployment since they are designed to verify the response from LLMs only after their entire generation (from the first to last tokens) is done. Further, we observe that once LLMs generate incorrect tokens early on, there is a higher likelihood that subsequent tokens will also be factually incorrect. To this end, in this work, we propose Streaming-VR (Streaming Verification and Refinement), a novel approach designed to enhance the efficiency of verification and refinement of LLM outputs. Specifically, the proposed Streaming-VR enables on-the-fly verification and correction of tokens as they are being generated, similar to a streaming process, ensuring that each subset of tokens is checked and refined in real-time by another LLM as the LLM constructs its response. Through comprehensive evaluations on multiple datasets, we demonstrate that our approach not only enhances the factual accuracy of LLMs, but also offers a more efficient solution compared to prior refinement methods.
comment: Preprint
☆ A Multi-Encoder Frozen-Decoder Approach for Fine-Tuning Large Language Models
Among parameter-efficient fine-tuning methods, freezing has emerged as a popular strategy for speeding up training, reducing catastrophic forgetting, and improving downstream performance. We investigate the impact of freezing the decoder in a multi-task setup comprising diverse natural language tasks, aiming to reduce deployment overhead and enhance portability to novel tasks. Our experiments, conducted by fine-tuning both individual and multi-task setups on the AlexaTM model, reveal that freezing decoders is highly effective for tasks with natural language outputs and mitigates catastrophic forgetting in multilingual tasks. However, we find that pairing frozen decoders with a larger model can effectively maintain or even enhance performance in structured and QA tasks, making it a viable strategy for a broader range of task types.
☆ STTS-EAD: Improving Spatio-Temporal Learning Based Time Series Prediction via
Handling anomalies is a critical preprocessing step in multivariate time series prediction. However, existing approaches that separate anomaly preprocessing from model training for multivariate time series prediction encounter significant limitations. Specifically, these methods fail to utilize auxiliary information crucial for identifying latent anomalies associated with spatiotemporal factors during the preprocessing stage. Instead, they rely solely on data distribution for anomaly detection, which can result in the incorrect processing of numerous samples that could otherwise contribute positively to model training. To address this, we propose STTS-EAD, an end-to-end method that seamlessly integrates anomaly detection into the training process of multivariate time series forecasting and aims to improve Spatio-Temporal learning based Time Series prediction via Embedded Anomaly Detection. Our proposed STTS-EAD leverages spatio-temporal information for forecasting and anomaly detection, with the two parts alternately executed and optimized for each other. To the best of our knowledge, STTS-EAD is the first to integrate anomaly detection and forecasting tasks in the training phase for improving the accuracy of multivariate time series forecasting. Extensive experiments on a public stock dataset and two real-world sales datasets from a renowned coffee chain enterprise show that our proposed method can effectively process detected anomalies in the training stage to improve forecasting performance in the inference stage and significantly outperform baselines.
comment: 11 pages
☆ Conformal mapping Coordinates Physics-Informed Neural Networks (CoCo-PINNs): learning neural networks for designing neutral inclusions
We focus on designing and solving the neutral inclusion problem via neural networks. The neutral inclusion problem has a long history in the theory of composite materials, and it is exceedingly challenging to identify the precise condition that precipitates a general-shaped inclusion into a neutral inclusion. Physics-informed neural networks (PINNs) have recently become a highly successful approach to addressing both forward and inverse problems associated with partial differential equations. We found that traditional PINNs perform inadequately when applied to the inverse problem of designing neutral inclusions with arbitrary shapes. In this study, we introduce a novel approach, Conformal mapping Coordinates Physics-Informed Neural Networks (CoCo-PINNs), which integrates complex analysis techniques into PINNs. This method exhibits strong performance in solving forward-inverse problems to construct neutral inclusions of arbitrary shapes in two dimensions, where the imperfect interface condition on the inclusion's boundary is modeled by training neural networks. Notably, we mathematically prove that training with a single linear field is sufficient to achieve neutrality for untrained linear fields in arbitrary directions, given a minor assumption. We demonstrate that CoCo-PINNs offer enhanced performances in terms of credibility, consistency, and stability.
☆ BioPose: Biomechanically-accurate 3D Pose Estimation from Monocular Videos
Recent advancements in 3D human pose estimation from single-camera images and videos have relied on parametric models, like SMPL. However, these models oversimplify anatomical structures, limiting their accuracy in capturing true joint locations and movements, which reduces their applicability in biomechanics, healthcare, and robotics. Biomechanically accurate pose estimation, on the other hand, typically requires costly marker-based motion capture systems and optimization techniques in specialized labs. To bridge this gap, we propose BioPose, a novel learning-based framework for predicting biomechanically accurate 3D human pose directly from monocular videos. BioPose includes three key components: a Multi-Query Human Mesh Recovery model (MQ-HMR), a Neural Inverse Kinematics (NeurIK) model, and a 2D-informed pose refinement technique. MQ-HMR leverages a multi-query deformable transformer to extract multi-scale fine-grained image features, enabling precise human mesh recovery. NeurIK treats the mesh vertices as virtual markers, applying a spatial-temporal network to regress biomechanically accurate 3D poses under anatomical constraints. To further improve 3D pose estimations, a 2D-informed refinement step optimizes the query tokens during inference by aligning the 3D structure with 2D pose observations. Experiments on benchmark datasets demonstrate that BioPose significantly outperforms state-of-the-art methods. Project website: \url{https://m-usamasaleem.github.io/publication/BioPose/BioPose.html}.
☆ Linearly Convergent Mixup Learning
Learning in the reproducing kernel Hilbert space (RKHS) such as the support vector machine has been recognized as a promising technique. It continues to be highly effective and competitive in numerous prediction tasks, particularly in settings where there is a shortage of training data or computational limitations exist. These methods are especially valued for their ability to work with small datasets and their interpretability. To address the issue of limited training data, mixup data augmentation, widely used in deep learning, has remained challenging to apply to learning in RKHS due to the generation of intermediate class labels. Although gradient descent methods handle these labels effectively, dual optimization approaches are typically not directly applicable. In this study, we present two novel algorithms that extend to a broader range of binary classification models. Unlike gradient-based approaches, our algorithms do not require hyperparameters like learning rates, simplifying their implementation and optimization. Both the number of iterations to converge and the computational cost per iteration scale linearly with respect to the dataset size. The numerical experiments demonstrate that our algorithms achieve faster convergence to the optimal solution compared to gradient descent approaches, and that mixup data augmentation consistently improves the predictive performance across various loss functions.
comment: none
☆ Transforming Indoor Localization: Advanced Transformer Architecture for NLOS Dominated Wireless Environments with Distributed Sensors
Indoor localization in challenging non-line-of-sight (NLOS) environments often leads to mediocre accuracy with traditional approaches. Deep learning (DL) has been applied to tackle these challenges; however, many DL approaches overlook computational complexity, especially for floating-point operations (FLOPs), making them unsuitable for resource-limited devices. Transformer-based models have achieved remarkable success in natural language processing (NLP) and computer vision (CV) tasks, motivating their use in wireless applications. However, their use in indoor localization remains nascent, and directly applying Transformers for indoor localization can be both computationally intensive and exhibit limitations in accuracy. To address these challenges, in this work, we introduce a novel tokenization approach, referred to as Sensor Snapshot Tokenization (SST), which preserves variable-specific representations of power delay profile (PDP) and enhances attention mechanisms by effectively capturing multi-variate correlation. Complementing this, we propose a lightweight Swish-Gated Linear Unit-based Transformer (L-SwiGLU Transformer) model, designed to reduce computational complexity without compromising localization accuracy. Together, these contributions mitigate the computational burden and dependency on large datasets, making Transformer models more efficient and suitable for resource-constrained scenarios. The proposed tokenization method enables the Vanilla Transformer to achieve a 90th percentile positioning error of 0.388 m in a highly NLOS indoor factory, surpassing conventional tokenization methods. The L-SwiGLU ViT further reduces the error to 0.355 m, achieving an 8.51% improvement. Additionally, the proposed model outperforms a 14.1 times larger model with a 46.13% improvement, underscoring its computational efficiency.
comment: The paper has been submitted to IEEE Transactions on Machine Learning in Communications and Networking
☆ Symmetry-Aware Generative Modeling through Learned Canonicalization
Generative modeling of symmetric densities has a range of applications in AI for science, from drug discovery to physics simulations. The existing generative modeling paradigm for invariant densities combines an invariant prior with an equivariant generative process. However, we observe that this technique is not necessary and has several drawbacks resulting from the limitations of equivariant networks. Instead, we propose to model a learned slice of the density so that only one representative element per orbit is learned. To accomplish this, we learn a group-equivariant canonicalization network that maps training samples to a canonical pose and train a non-equivariant generative model over these canonicalized samples. We implement this idea in the context of diffusion models. Our preliminary experimental results on molecular modeling are promising, demonstrating improved sample quality and faster inference time.
comment: NeurReps 2024 Workshop Version
☆ BMIP: Bi-directional Modality Interaction Prompt Learning for VLM
Vision-language models (VLMs) have exhibited remarkable generalization capabilities, and prompt learning for VLMs has attracted great attention for the ability to adapt pre-trained VLMs to specific downstream tasks. However, existing studies mainly focus on single-modal prompts or uni-directional modality interaction, overlooking the powerful alignment effects resulting from the interaction between the vision and language modalities. To this end, we propose a novel prompt learning method called $\underline{\textbf{B}}i-directional \underline{\textbf{M}}odality \underline{\textbf{I}}nteraction \underline{\textbf{P}}rompt (BMIP)$, which dynamically weights bi-modal information through learning the information of the attention layer, enhancing trainability and inter-modal consistency compared to simple information aggregation methods. To evaluate the effectiveness of prompt learning methods, we propose a more realistic evaluation paradigm called open-world generalization complementing the widely adopted cross-dataset transfer and domain generalization tasks. Comprehensive experiments on various datasets reveal that BMIP not only outperforms current state-of-the-art methods across all three evaluation paradigms but is also flexible enough to be combined with other prompt-based methods for consistent performance enhancement.
☆ Quantifying the Importance of Data Alignment in Downstream Model Performance
Contrary to the conventional emphasis on dataset size, we explore the role of data alignment -- an often overlooked aspect of data quality -- in training capable Large Language Models (LLMs). To do so, we use the Task2Vec-based alignment coefficient, a quantitative measure of the similarity between two datasets, to quantify the impact of alignment between training data and evaluation data on downstream performance. In particular, we conduct controlled \textit{interventional} experiments for two settings: 1. the impact of increased alignment coefficients between various pre-training (pt) against evaluation datasets, and 2. the impact of increased alignment coefficients between domain specific fine-tuning (ft) against domain specific evaluation. The domain specific task we explore is Autoformalization -- the machine translation task between natural language and code for formal verification. In both settings, we find a strong, predictable negative correlation between the alignment coefficient of a model's training and evaluation data and the model's loss/perplexity on the respective downstream task. These findings suggest a re-evaluation of LLM training approaches, demonstrating the relevance of data alignment compared to data quantity, especially in specialized downstream tasks such as Autoformalization.
☆ FLAVARS: A Multimodal Foundational Language and Vision Alignment Model for Remote Sensing
Remote sensing imagery is dense with objects and contextual visual information. There is a recent trend to combine paired satellite images and text captions for pretraining performant encoders for downstream tasks. However, while contrastive image-text methods like CLIP enable vision-language alignment and zero-shot classification ability, vision-only downstream performance tends to degrade compared to image-only pretraining, such as MAE. In this paper, we propose FLAVARS, a pretraining method that combines the best of both contrastive learning and masked modeling, along with geospatial alignment via contrastive location encoding. We find that FLAVARS significantly outperforms a baseline of SkyCLIP for vision-only tasks such as KNN classification and semantic segmentation, +6\% mIOU on SpaceNet1, while retaining the ability to perform zero-shot classification, unlike MAE pretrained methods.
☆ Time series forecasting for multidimensional telemetry data using GAN and BiLSTM in a Digital Twin
The research related to digital twins has been increasing in recent years. Besides the mirroring of the physical word into the digital, there is the need of providing services related to the data collected and transferred to the virtual world. One of these services is the forecasting of physical part future behavior, that could lead to applications, like preventing harmful events or designing improvements to get better performance. One strategy used to predict any system operation it is the use of time series models like ARIMA or LSTM, and improvements were implemented using these algorithms. Recently, deep learning techniques based on generative models such as Generative Adversarial Networks (GANs) have been proposed to create time series and the use of LSTM has gained more relevance in time series forecasting, but both have limitations that restrict the forecasting results. Another issue found in the literature is the challenge of handling multivariate environments/applications in time series generation. Therefore, new methods need to be studied in order to fill these gaps and, consequently, provide better resources for creating useful digital twins. In this proposal, it is going to be studied the integration of a BiLSTM layer with a time series obtained by GAN in order to improve the forecasting of all the features provided by the dataset in terms of accuracy and, consequently, improving behaviour prediction.
☆ Head Motion Degrades Machine Learning Classification of Alzheimer's Disease from Positron Emission Tomography
Brain positron emission tomography (PET) imaging is broadly used in research and clinical routines to study, diagnose, and stage Alzheimer's disease (AD). However, its potential cannot be fully exploited yet due to the lack of portable motion correction solutions, especially in clinical settings. Head motion during data acquisition has indeed been shown to degrade image quality and induces tracer uptake quantification error. In this study, we demonstrate that it also biases machine learning-based AD classification. We start by proposing a binary classification algorithm solely based on PET images. We find that it reaches a high accuracy in classifying motion corrected images into cognitive normal or AD. We demonstrate that the classification accuracy substantially decreases when images lack motion correction, thereby limiting the algorithm's effectiveness and biasing image interpretation. We validate these findings in cohorts of 128 $^{11}$C-UCB-J and 173 $^{18}$F-FDG scans, two tracers highly relevant to the study of AD. Classification accuracies decreased by 10% and 5% on 20 $^{18}$F-FDG and 20 $^{11}$C-UCB-J testing cases, respectively. Our findings underscore the critical need for efficient motion correction methods to make the most of the diagnostic capabilities of PET-based machine learning.
comment: 5 pages
☆ Large Language Models For Text Classification: Case Study And Comprehensive Review
Unlocking the potential of Large Language Models (LLMs) in data classification represents a promising frontier in natural language processing. In this work, we evaluate the performance of different LLMs in comparison with state-of-the-art deep-learning and machine-learning models, in two different classification scenarios: i) the classification of employees' working locations based on job reviews posted online (multiclass classification), and 2) the classification of news articles as fake or not (binary classification). Our analysis encompasses a diverse range of language models differentiating in size, quantization, and architecture. We explore the impact of alternative prompting techniques and evaluate the models based on the weighted F1-score. Also, we examine the trade-off between performance (F1-score) and time (inference response time) for each language model to provide a more nuanced understanding of each model's practical applicability. Our work reveals significant variations in model responses based on the prompting strategies. We find that LLMs, particularly Llama3 and GPT-4, can outperform traditional methods in complex classification tasks, such as multiclass classification, though at the cost of longer inference times. In contrast, simpler ML models offer better performance-to-time trade-offs in simpler binary classification tasks.
Rate-In: Information-Driven Adaptive Dropout Rates for Improved Inference-Time Uncertainty Estimation
Accurate uncertainty estimation is crucial for deploying neural networks in risk-sensitive applications such as medical diagnosis. Monte Carlo Dropout is a widely used technique for approximating predictive uncertainty by performing stochastic forward passes with dropout during inference. However, using static dropout rates across all layers and inputs can lead to suboptimal uncertainty estimates, as it fails to adapt to the varying characteristics of individual inputs and network layers. Existing approaches optimize dropout rates during training using labeled data, resulting in fixed inference-time parameters that cannot adjust to new data distributions, compromising uncertainty estimates in Monte Carlo simulations. In this paper, we propose Rate-In, an algorithm that dynamically adjusts dropout rates during inference by quantifying the information loss induced by dropout in each layer's feature maps. By treating dropout as controlled noise injection and leveraging information-theoretic principles, Rate-In adapts dropout rates per layer and per input instance without requiring ground truth labels. By quantifying the functional information loss in feature maps, we adaptively tune dropout rates to maintain perceptual quality across diverse medical imaging tasks and architectural configurations. Our extensive empirical study on synthetic data and real-world medical imaging tasks demonstrates that Rate-In improves calibration and sharpens uncertainty estimates compared to fixed or heuristic dropout rates without compromising predictive performance. Rate-In offers a practical, unsupervised, inference-time approach to optimizing dropout for more reliable predictive uncertainty estimation in critical applications.
comment: Updated author affiliation
♻ ☆ Efficient Distribution Matching of Representations via Noise-Injected Deep InfoMax
Deep InfoMax (DIM) is a well-established method for self-supervised representation learning (SSRL) based on maximization of the mutual information between the input and the output of a deep neural network encoder. Despite the DIM and contrastive SSRL in general being well-explored, the task of learning representations conforming to a specific distribution (i.e., distribution matching, DM) is still under-addressed. Motivated by the importance of DM to several downstream tasks (including generative modeling, disentanglement, outliers detection and other), we enhance DIM to enable automatic matching of learned representations to a selected prior distribution. To achieve this, we propose injecting an independent noise into the normalized outputs of the encoder, while keeping the same InfoMax training objective. We show that such modification allows for learning uniformly and normally distributed representations, as well as representations of other absolutely continuous distributions. Our approach is tested on various downstream tasks. The results indicate a moderate trade-off between the performance on the downstream tasks and quality of DM.
comment: 25 pages, 7 fugures
♻ ☆ CriSPO: Multi-Aspect Critique-Suggestion-guided Automatic Prompt Optimization for Text Generation AAAI-2025
Existing automatic prompt engineering methods are typically designed for discriminative tasks, where new task prompts are iteratively refined with limited feedback from a single metric reflecting a single aspect. However, these approaches are suboptimal for generative tasks, which require more nuanced guidance beyond a single numeric metric to improve the prompt and optimize multiple aspects of the generated text. To address these challenges, we propose a novel multi-aspect Critique-Suggestion-guided automatic Prompt Optimization (CriSPO) approach. CriSPO introduces a critique-suggestion module as its core component. This module spontaneously discovers aspects, and compares generated and reference texts across these aspects, providing specific suggestions for prompt modification. These clear critiques and actionable suggestions guide a receptive optimizer module to make more substantial changes, exploring a broader and more effective search space. To further improve CriSPO with multi-metric optimization, we introduce an Automatic Suffix Tuning (AST) extension to enhance the performance of task prompts across multiple metrics. We evaluate CriSPO on 4 state-of-the-art LLMs across 4 summarization and 5 QA datasets. Extensive experiments show 3-4% ROUGE score improvement on summarization and substantial improvement of various metrics on QA. Code available at https://github.com/amazon-science/crispo
comment: Accepted to AAAI-2025
♻ ☆ Automated Detection and Analysis of Minor Deformations in Flat Walls Due to Railway Vibrations Using LiDAR and Machine Learning
This study introduces an advanced methodology for automatically identifying minor deformations in flat walls caused by vibrations from nearby railway tracks. It leverages high-density Terrestrial Laser Scanner (TLS) LiDAR surveys and AI/ML techniques to collect and analyze data. The scan data is processed into a detailed point cloud, which is segmented to distinguish ground points, trees, buildings, and other objects. The analysis focuses on identifying sections along flat walls and estimating their deformations relative to the ground orientation. Findings from the study, conducted at the RGIPT campus, reveal significant deformations in walls close to the railway corridor, with the highest deformations ranging from 7 to 8 cm and an average of 3 to 4 cm. In contrast, walls further from the corridor show negligible deformations. The developed automated process for feature extraction and deformation monitoring demonstrates potential for structural health monitoring. By integrating LiDAR data with machine learning, the methodology provides an efficient system for identifying and analyzing structural deformations, highlighting the importance of continuous monitoring for ensuring structural integrity and public safety in urban infrastructure. This approach represents a substantial advancement in automated feature extraction and deformation analysis, contributing to more effective management of urban infrastructure.
comment: I am requesting the withdrawal of my paper due to the need for significant revisions to ensure the accuracy and integrity of the presented findings
♻ ☆ Language-Agnostic Modeling of Source Reliability on Wikipedia
Over the last few years, content verification through reliable sources has become a fundamental need to combat disinformation. Here, we present a language-agnostic model designed to assess the reliability of sources across multiple language editions of Wikipedia. Utilizing editorial activity data, the model evaluates source reliability within different articles of varying controversiality such as Climate Change, COVID-19, History, Media, and Biology topics. Crafting features that express domain usage across articles, the model effectively predicts source reliability, achieving an F1 Macro score of approximately 0.80 for English and other high-resource languages. For mid-resource languages, we achieve 0.65 while the performance of low-resource languages varies; in all cases, the time the domain remains present in the articles (which we dub as permanence) is one of the most predictive features. We highlight the challenge of maintaining consistent model performance across languages of varying resource levels and demonstrate that adapting models from higher-resource languages can improve performance. This work contributes not only to Wikipedia's efforts in ensuring content verifiability but in ensuring reliability across diverse user-generated content in various language communities.
♻ ☆ A Comprehensive Survey of Foundation Models in Medicine
Foundation models (FMs) are large-scale deep learning models that are developed using large datasets and self-supervised learning methods. These models serve as a base for different downstream tasks, including healthcare. FMs have been adopted with great success across various domains within healthcare. Existing healthcare-based surveys have not yet included all of these domains. Therefore, we provide a detailed survey of FMs in healthcare. We focus on the history, learning strategies, flagship models, applications, and challenges of FMs. We explore how FMs such as the BERT and GPT families are reshaping various healthcare domains, including clinical large language models, medical image analysis, and omics. Furthermore, we provide a detailed taxonomy of healthcare applications facilitated by FMs, such as clinical NLP, medical computer vision, graph learning, and other biology-related tasks. Despite the promising opportunities FMs provide, they also have several associated challenges, which are explained in detail. We also outline open research issues and potential lessons learned to provide researchers and practitioners with insights into the capabilities of FMs in healthcare to advance their deployment and mitigate associated risks.
comment: Currently under review in IEEE REVIEWS IN BIOMEDICAL ENGINEERING
♻ ☆ Pareto Set Learning for Multi-Objective Reinforcement Learning AAAI 2025
Multi-objective decision-making problems have emerged in numerous real-world scenarios, such as video games, navigation and robotics. Considering the clear advantages of Reinforcement Learning (RL) in optimizing decision-making processes, researchers have delved into the development of Multi-Objective RL (MORL) methods for solving multi-objective decision problems. However, previous methods either cannot obtain the entire Pareto front, or employ only a single policy network for all the preferences over multiple objectives, which may not produce personalized solutions for each preference. To address these limitations, we propose a novel decomposition-based framework for MORL, Pareto Set Learning for MORL (PSL-MORL), that harnesses the generation capability of hypernetwork to produce the parameters of the policy network for each decomposition weight, generating relatively distinct policies for various scalarized subproblems with high efficiency. PSL-MORL is a general framework, which is compatible for any RL algorithm. The theoretical result guarantees the superiority of the model capacity of PSL-MORL and the optimality of the obtained policy network. Through extensive experiments on diverse benchmarks, we demonstrate the effectiveness of PSL-MORL in achieving dense coverage of the Pareto front, significantly outperforming state-of-the-art MORL methods in the hypervolume and sparsity indicators.
comment: AAAI 2025 Accept
♻ ☆ Feedback-driven object detection and iterative model improvement
Automated object detection has become increasingly valuable across diverse applications, yet efficient, high-quality annotation remains a persistent challenge. In this paper, we present the development and evaluation of a platform designed to interactively improve object detection models. The platform allows uploading and annotating images as well as fine-tuning object detection models. Users can then manually review and refine annotations, further creating improved snapshots that are used for automatic object detection on subsequent image uploads - a process we refer to as semi-automatic annotation resulting in a significant gain in annotation efficiency. Whereas iterative refinement of model results to speed up annotation has become common practice, we are the first to quantitatively evaluate its benefits with respect to time, effort, and interaction savings. Our experimental results show clear evidence for a significant time reduction of up to 53% for semi-automatic compared to manual annotation. Importantly, these efficiency gains did not compromise annotation quality, while matching or occasionally even exceeding the accuracy of manual annotations. These findings demonstrate the potential of our lightweight annotation platform for creating high-quality object detection datasets and provide best practices to guide future development of annotation platforms. The platform is open-source, with the frontend and backend repositories available on GitHub (https://github.com/ml-lab-htw/iterative-annotate). To support the understanding of our labeling process, we have created an explanatory video demonstrating the methodology using microscopy images of E. coli bacteria as an example. The video is available on YouTube (https://www.youtube.com/watch?v=CM9uhE8NN5E).
comment: AI4EA24
♻ ☆ ORFormer: Occlusion-Robust Transformer for Accurate Facial Landmark Detection WACV 2025
Although facial landmark detection (FLD) has gained significant progress, existing FLD methods still suffer from performance drops on partially non-visible faces, such as faces with occlusions or under extreme lighting conditions or poses. To address this issue, we introduce ORFormer, a novel transformer-based method that can detect non-visible regions and recover their missing features from visible parts. Specifically, ORFormer associates each image patch token with one additional learnable token called the messenger token. The messenger token aggregates features from all but its patch. This way, the consensus between a patch and other patches can be assessed by referring to the similarity between its regular and messenger embeddings, enabling non-visible region identification. Our method then recovers occluded patches with features aggregated by the messenger tokens. Leveraging the recovered features, ORFormer compiles high-quality heatmaps for the downstream FLD task. Extensive experiments show that our method generates heatmaps resilient to partial occlusions. By integrating the resultant heatmaps into existing FLD methods, our method performs favorably against the state of the arts on challenging datasets such as WFLW and COFW.
comment: WACV 2025 Project Link: https://ben0919.github.io/ORFormer/
♻ ☆ WINE: Wavelet-Guided GAN Inversion and Editing for High-Fidelity Refinement
Recent advanced GAN inversion models aim to convey high-fidelity information from original images to generators through methods using generator tuning or high-dimensional feature learning. Despite these efforts, accurately reconstructing image-specific details remains as a challenge due to the inherent limitations both in terms of training and structural aspects, leading to a bias towards low-frequency information. In this paper, we look into the widely used pixel loss in GAN inversion, revealing its predominant focus on the reconstruction of low-frequency features. We then propose WINE, a Wavelet-guided GAN Inversion aNd Editing model, which transfers the high-frequency information through wavelet coefficients via newly proposed wavelet loss and wavelet fusion scheme. Notably, WINE is the first attempt to interpret GAN inversion in the frequency domain. Our experimental results showcase the precision of WINE in preserving high-frequency details and enhancing image quality. Even in editing scenarios, WINE outperforms existing state-of-the-art GAN inversion models with a fine balance between editability and reconstruction quality.
♻ ☆ AttriBoT: A Bag of Tricks for Efficiently Approximating Leave-One-Out Context Attribution
The influence of contextual input on the behavior of large language models (LLMs) has prompted the development of context attribution methods that aim to quantify each context span's effect on an LLM's generations. The leave-one-out (LOO) error, which measures the change in the likelihood of the LLM's response when a given span of the context is removed, provides a principled way to perform context attribution, but can be prohibitively expensive to compute for large models. In this work, we introduce AttriBoT, a series of novel techniques for efficiently computing an approximation of the LOO error for context attribution. Specifically, AttriBoT uses cached activations to avoid redundant operations, performs hierarchical attribution to reduce computation, and emulates the behavior of large target models with smaller proxy models. Taken together, AttriBoT can provide a >300x speedup while remaining more faithful to a target model's LOO error than prior context attribution methods. This stark increase in performance makes computing context attributions for a given response 30x faster than generating the response itself, empowering real-world applications that require computing attributions at scale. We release a user-friendly and efficient implementation of AttriBoT to enable efficient LLM interpretability as well as encourage future development of efficient context attribution methods.
comment: 29 pages, 11 figures
♻ ☆ Electricity Price Prediction Using Multi-Kernel Gaussian Process Regression Combined with Kernel-Based Support Vector Regression
This paper presents a new hybrid model for predicting German electricity prices. The algorithm is based on combining Gaussian Process Regression (GPR) and Support Vector Regression (SVR). While GPR is a competent model for learning the stochastic pattern within the data and interpolation, its performance for out-of-sample data is not very promising. By choosing a suitable data-dependent covariance function, we can enhance the performance of GPR for the tested German hourly power prices. However, since the out-of-sample prediction depends on the training data, the prediction is vulnerable to noise and outliers. To overcome this issue, a separate prediction is made using SVR, which applies margin-based optimization, having an advantage in dealing with non-linear processes and outliers, since only certain necessary points (support vectors) in the training data are responsible for regression. Both individual predictions are later combined using the performance-based weight assignment method. A test on historic German power prices shows that this approach outperforms its chosen benchmarks such as the autoregressive exogenous model, the naive approach, as well as the long short-term memory approach of prediction.
♻ ☆ Set-based Neural Network Encoding Without Weight Tying
We propose a neural network weight encoding method for network property prediction that utilizes set-to-set and set-to-vector functions to efficiently encode neural network parameters. Our approach is capable of encoding neural networks in a model zoo of mixed architecture and different parameter sizes as opposed to previous approaches that require custom encoding models for different architectures. Furthermore, our \textbf{S}et-based \textbf{N}eural network \textbf{E}ncoder (SNE) takes into consideration the hierarchical computational structure of neural networks. To respect symmetries inherent in network weight space, we utilize Logit Invariance to learn the required minimal invariance properties. Additionally, we introduce a \textit{pad-chunk-encode} pipeline to efficiently encode neural network layers that is adjustable to computational and memory constraints. We also introduce two new tasks for neural network property prediction: cross-dataset and cross-architecture. In cross-dataset property prediction, we evaluate how well property predictors generalize across model zoos trained on different datasets but of the same architecture. In cross-architecture property prediction, we evaluate how well property predictors transfer to model zoos of different architecture not seen during training. We show that SNE outperforms the relevant baselines on standard benchmarks.
comment: 23 pages
♻ ☆ Approximation Rates in Fréchet Metrics: Barron Spaces, Paley-Wiener Spaces, and Fourier Multipliers
Operator learning is a recent development in the simulation of Partial Differential Equations (PDEs) by means of neural networks. The idea behind this approach is to learn the behavior of an operator, such that the resulting neural network is an (approximate) mapping in infinite-dimensional spaces that is capable of (approximately) simulating the solution operator governed by the PDE. In our work, we study some general approximation capabilities for linear differential operators by approximating the corresponding symbol in the Fourier domain. Analogous to the structure of the class of H\"ormander-Symbols, we consider the approximation with respect to a topology that is induced by a sequence of semi-norms. In that sense, we measure the approximation error in terms of a Fr\'echet metric, and our main result identifies sufficient conditions for achieving a predefined approximation error. Secondly, we then focus on a natural extension of our main theorem, in which we manage to reduce the assumptions on the sequence of semi-norms. Based on existing approximation results for the exponential spectral Barron space, we then present a concrete example of symbols that can be approximated well.
comment: Minor revision
♻ ☆ Towards Federated Graph Learning in One-shot Communication
Federated Graph Learning (FGL) has emerged as a promising paradigm for breaking data silos among distributed private graphs. In practical scenarios involving heterogeneous distributed graph data, personalized Federated Graph Learning (pFGL) aims to enhance model utility by training personalized models tailored to client needs. However, existing pFGL methods often require numerous communication rounds under heterogeneous graphs, leading to significant communication overhead and security concerns. While One-shot Federated Learning (OFL) enables collaboration in a single round, existing OFL methods are designed for image-centric tasks and ineffective for graph data, leaving a critical gap in the field. Additionally, personalized models derived from existing methods suffer from bias, failing to effectively generalize to the minority. To address these challenges, we propose the first $\textbf{O}$ne-shot $\textbf{p}$ersonalized $\textbf{F}$ederated $\textbf{G}$raph $\textbf{L}$earning method ($\textbf{O-pFGL}$) for node classification, compatible with Secure Aggregation protocols for privacy preservation. Specifically, for effective graph learning in one communication round, our method estimates and aggregates class-wise feature distribution statistics to construct a global pseudo-graph on the server, facilitating the training of a global graph model. To mitigate bias, we introduce a two-stage personalized training approach that adaptively balances local personal information and global insights from the pseudo-graph, improving both personalization and generalization. Extensive experiments on 12 multi-scale graph datasets demonstrate that our method significantly outperforms state-of-the-art baselines across various settings.
comment: Work in progress
♻ ☆ Dynamic Sub-graph Distillation for Robust Semi-supervised Continual Learning
Continual learning (CL) has shown promising results and comparable performance to learning at once in a fully supervised manner. However, CL strategies typically require a large number of labeled samples, making their real-life deployment challenging. In this work, we focus on semi-supervised continual learning (SSCL), where the model progressively learns from partially labeled data with unknown categories. We provide a comprehensive analysis of SSCL and demonstrate that unreliable distributions of unlabeled data lead to unstable training and refinement of the progressing stages. This problem severely impacts the performance of SSCL. To address the limitations, we propose a novel approach called Dynamic Sub-Graph Distillation (DSGD) for semi-supervised continual learning, which leverages both semantic and structural information to achieve more stable knowledge distillation on unlabeled data and exhibit robustness against distribution bias. Firstly, we formalize a general model of structural distillation and design a dynamic graph construction for the continual learning progress. Next, we define a structure distillation vector and design a dynamic sub-graph distillation algorithm, which enables end-to-end training and adaptability to scale up tasks. The entire proposed method is adaptable to various CL methods and supervision settings. Finally, experiments conducted on three datasets CIFAR10, CIFAR100, and ImageNet-100, with varying supervision ratios, demonstrate the effectiveness of our proposed approach in mitigating the catastrophic forgetting problem in semi-supervised continual learning scenarios.
♻ ☆ Balanced Neural ODEs: nonlinear model order reduction and Koopman operator approximations
Variational Autoencoders (VAEs) are a powerful framework for learning latent representations of reduced dimensionality, while Neural ODEs excel in learning transient system dynamics. This work combines the strengths of both to generate fast surrogate models with adjustable complexity reacting on time-varying inputs signals. By leveraging the VAE's dimensionality reduction using a nonhierarchical prior, our method adaptively assigns stochastic noise, naturally complementing known NeuralODE training enhancements and enabling probabilistic time series modeling. We show that standard Latent ODEs struggle with dimensionality reduction in systems with time-varying inputs. Our approach mitigates this by continuously propagating variational parameters through time, establishing fixed information channels in latent space. This results in a flexible and robust method that can learn different system complexities, e.g. deep neural networks or linear matrices. Hereby, it enables efficient approximation of the Koopman operator without the need for predefining its dimensionality. As our method balances dimensionality reduction and reconstruction accuracy, we call it Balanced Neural ODE (B-NODE). We demonstrate the effectiveness of this methods on several academic and real-world test cases, e.g. a power plant or MuJoCo data.
comment: Conference paper under review, after revision
♻ ☆ Spurious Feature Eraser: Stabilizing Test-Time Adaptation for Vision-Language Foundation Model
Vision-language foundation models have exhibited remarkable success across a multitude of downstream tasks due to their scalability on extensive image-text paired data. However, these models also display significant limitations when applied to downstream tasks, such as fine-grained image classification, as a result of ``decision shortcuts'' that hinder their generalization capabilities. In this work, we find that the CLIP model possesses a rich set of features, encompassing both \textit{desired invariant causal features} and \textit{undesired decision shortcuts}. Moreover, the underperformance of CLIP on downstream tasks originates from its inability to effectively utilize pre-trained features in accordance with specific task requirements. To address this challenge, we propose a simple yet effective method, Spurious Feature Eraser (SEraser), to alleviate the decision shortcuts by erasing the spurious features. Specifically, we introduce a test-time prompt tuning paradigm that optimizes a learnable prompt, thereby compelling the model to exploit invariant features while disregarding decision shortcuts during the inference phase. The proposed method effectively alleviates excessive dependence on potentially misleading spurious information. We conduct comparative analysis of the proposed method against various approaches which validates the significant superiority.
♻ ☆ ImagiNet: A Multi-Content Benchmark for Synthetic Image Detection AAAI 2025
Recent generative models produce images with a level of authenticity that makes them nearly indistinguishable from real photos and artwork. Potential harmful use cases of these models, necessitate the creation of robust synthetic image detectors. However, current datasets in the field contain generated images with questionable quality or have examples from one predominant content type which leads to poor generalizability of the underlying detectors. We find that the curation of a balanced amount of high-resolution generated images across various content types is crucial for the generalizability of detectors, and introduce ImagiNet, a dataset of 200K examples, spanning four categories: photos, paintings, faces, and miscellaneous. Synthetic images in ImagiNet are produced with both open-source and proprietary generators, whereas real counterparts for each content type are collected from public datasets. The structure of ImagiNet allows for a two-track evaluation system: i) classification as real or synthetic and ii) identification of the generative model. To establish a strong baseline, we train a ResNet-50 model using a self-supervised contrastive objective (SelfCon) for each track which achieves evaluation AUC of up to 0.99 and balanced accuracy ranging from 86% to 95%, even under conditions that involve compression and resizing. The provided model is generalizable enough to achieve zero-shot state-of-the-art performance on previous synthetic detection benchmarks. We provide ablations to demonstrate the importance of content types and publish code and data.
comment: Workshop on Datasets and Evaluators of AI Safety, AAAI 2025
♻ ☆ Audio-Agent: Leveraging LLMs For Audio Generation, Editing and Composition
We introduce Audio-Agent, a multimodal framework for audio generation, editing and composition based on text or video inputs. Conventional approaches for text-to-audio (TTA) tasks often make single-pass inferences from text descriptions. While straightforward, this design struggles to produce high-quality audio when given complex text conditions. In our method, we utilize a pre-trained TTA diffusion network as the audio generation agent to work in tandem with GPT-4, which decomposes the text condition into atomic, specific instructions and calls the agent for audio generation. In doing so, Audio-Agent can generate high-quality audio that is closely aligned with the provided text or video exhibiting complex and multiple events, while supporting variable-length and variable-volume generation. For video-to-audio (VTA) tasks, most existing methods require training a timestamp detector to synchronize video events with the generated audio, a process that can be tedious and time-consuming. Instead, we propose a simpler approach by fine-tuning a pre-trained Large Language Model (LLM), e.g., Gemma2-2B-it, to obtain both semantic and temporal conditions that bridge the video and audio modality. Consequently, our framework contributes a comprehensive solution for both TTA and VTA tasks without substantial computational overhead in training.
♻ ☆ Correlation-Aware Graph Convolutional Networks for Multi-Label Node Classification KDD2025
Multi-label node classification is an important yet under-explored domain in graph mining as many real-world nodes belong to multiple categories rather than just a single one. Although a few efforts have been made by utilizing Graph Convolution Networks (GCNs) to learn node representations and model correlations between multiple labels in the embedding space, they still suffer from the ambiguous feature and ambiguous topology induced by multiple labels, which reduces the credibility of the messages delivered in graphs and overlooks the label correlations on graph data. Therefore, it is crucial to reduce the ambiguity and empower the GCNs for accurate classification. However, this is quite challenging due to the requirement of retaining the distinctiveness of each label while fully harnessing the correlation between labels simultaneously. To address these issues, in this paper, we propose a Correlation-aware Graph Convolutional Network (CorGCN) for multi-label node classification. By introducing a novel Correlation-Aware Graph Decomposition module, CorGCN can learn a graph that contains rich label-correlated information for each label. It then employs a Correlation-Enhanced Graph Convolution to model the relationships between labels during message passing to further bolster the classification process. Extensive experiments on five datasets demonstrate the effectiveness of our proposed CorGCN.
comment: 12 pages, accepted by KDD2025
♻ ☆ A Random Matrix Approach to Low-Multilinear-Rank Tensor Approximation
This work presents a comprehensive understanding of the estimation of a planted low-rank signal from a general spiked tensor model near the computational threshold. Relying on standard tools from the theory of large random matrices, we characterize the large-dimensional spectral behavior of the unfoldings of the data tensor and exhibit relevant signal-to-noise ratios governing the detectability of the principal directions of the signal. These results allow to accurately predict the reconstruction performance of truncated multilinear SVD (MLSVD) in the non-trivial regime. This is particularly important since it serves as an initialization of the higher-order orthogonal iteration (HOOI) scheme, whose convergence to the best low-multilinear-rank approximation depends entirely on its initialization. We give a sufficient condition for the convergence of HOOI and show that the number of iterations before convergence tends to $1$ in the large-dimensional limit.
♻ ☆ Fast, Scale-Adaptive, and Uncertainty-Aware Downscaling of Earth System Model Fields with Generative Machine Learning
Accurate and high-resolution Earth system model (ESM) simulations are essential to assess the ecological and socio-economic impacts of anthropogenic climate change, but are computationally too expensive to be run at sufficiently high spatial resolution. Recent machine learning approaches have shown promising results in downscaling ESM simulations, outperforming state-of-the-art statistical approaches. However, existing methods require computationally costly retraining for each ESM and extrapolate poorly to climates unseen during training. We address these shortcomings by learning a consistency model (CM) that efficiently and accurately downscales arbitrary ESM simulations without retraining in a zero-shot manner. Our approach yields probabilistic downscaled fields at a resolution only limited by the observational reference data. We show that the CM outperforms state-of-the-art diffusion models at a fraction of computational cost while maintaining high controllability on the downscaling task. Further, our method generalizes to climate states unseen during training without explicitly formulated physical constraints.
♻ ☆ Learning Symmetries via Weight-Sharing with Doubly Stochastic Tensors
Group equivariance has emerged as a valuable inductive bias in deep learning, enhancing generalization, data efficiency, and robustness. Classically, group equivariant methods require the groups of interest to be known beforehand, which may not be realistic for real-world data. Additionally, baking in fixed group equivariance may impose overly restrictive constraints on model architecture. This highlights the need for methods that can dynamically discover and apply symmetries as soft constraints. For neural network architectures, equivariance is commonly achieved through group transformations of a canonical weight tensor, resulting in weight sharing over a given group $G$. In this work, we propose to learn such a weight-sharing scheme by defining a collection of learnable doubly stochastic matrices that act as soft permutation matrices on canonical weight tensors, which can take regular group representations as a special case. This yields learnable kernel transformations that are jointly optimized with downstream tasks. We show that when the dataset exhibits strong symmetries, the permutation matrices will converge to regular group representations and our weight-sharing networks effectively become regular group convolutions. Additionally, the flexibility of the method enables it to effectively pick up on partial symmetries.
comment: 19 pages, 14 figures, 4 tables
♻ ☆ Scalable and Resource-Efficient Second-Order Federated Learning via Over-the-Air Aggregation
Second-order federated learning (FL) algorithms offer faster convergence than their first-order counterparts by leveraging curvature information. However, they are hindered by high computational and storage costs, particularly for large-scale models. Furthermore, the communication overhead associated with large models and digital transmission exacerbates these challenges, causing communication bottlenecks. In this work, we propose a scalable second-order FL algorithm using a sparse Hessian estimate and leveraging over-the-air aggregation, making it feasible for larger models. Our simulation results demonstrate more than $67\%$ of communication resources and energy savings compared to other first and second-order baselines.
comment: 6 pages, 1 figure, 4 subfigures, letter
♻ ☆ Rethinking Decoders for Transformer-based Semantic Segmentation: A Compression Perspective NeurIPS2024
State-of-the-art methods for Transformer-based semantic segmentation typically adopt Transformer decoders that are used to extract additional embeddings from image embeddings via cross-attention, refine either or both types of embeddings via self-attention, and project image embeddings onto the additional embeddings via dot-product. Despite their remarkable success, these empirical designs still lack theoretical justifications or interpretations, thus hindering potentially principled improvements. In this paper, we argue that there are fundamental connections between semantic segmentation and compression, especially between the Transformer decoders and Principal Component Analysis (PCA). From such a perspective, we derive a white-box, fully attentional DEcoder for PrIncipled semantiC segemenTation (DEPICT), with the interpretations as follows: 1) the self-attention operator refines image embeddings to construct an ideal principal subspace that aligns with the supervision and retains most information; 2) the cross-attention operator seeks to find a low-rank approximation of the refined image embeddings, which is expected to be a set of orthonormal bases of the principal subspace and corresponds to the predefined classes; 3) the dot-product operation yields compact representation for image embeddings as segmentation masks. Experiments conducted on dataset ADE20K find that DEPICT consistently outperforms its black-box counterpart, Segmenter, and it is light weight and more robust.
comment: NeurIPS2024. Code:https://github.com/QishuaiWen/DEPICT/
♻ ☆ GenSafe: A Generalizable Safety Enhancer for Safe Reinforcement Learning Algorithms Based on Reduced Order Markov Decision Process Model
Safe Reinforcement Learning (SRL) aims to realize a safe learning process for Deep Reinforcement Learning (DRL) algorithms by incorporating safety constraints. However, the efficacy of SRL approaches often relies on accurate function approximations, which are notably challenging to achieve in the early learning stages due to data insufficiency. To address this issue, we introduce in this work a novel Generalizable Safety enhancer (GenSafe) that is able to overcome the challenge of data insufficiency and enhance the performance of SRL approaches. Leveraging model order reduction techniques, we first propose an innovative method to construct a Reduced Order Markov Decision Process (ROMDP) as a low-dimensional approximator of the original safety constraints. Then, by solving the reformulated ROMDP-based constraints, GenSafe refines the actions of the agent to increase the possibility of constraint satisfaction. Essentially, GenSafe acts as an additional safety layer for SRL algorithms. We evaluate GenSafe on multiple SRL approaches and benchmark problems. The results demonstrate its capability to improve safety performance, especially in the early learning phases, while maintaining satisfactory task performance. Our proposed GenSafe not only offers a novel measure to augment existing SRL methods but also shows broad compatibility with various SRL algorithms, making it applicable to a wide range of systems and SRL problems.
♻ ☆ Fair CoVariance Neural Networks
Covariance-based data processing is widespread across signal processing and machine learning applications due to its ability to model data interconnectivities and dependencies. However, harmful biases in the data may become encoded in the sample covariance matrix and cause data-driven methods to treat different subpopulations unfairly. Existing works such as fair principal component analysis (PCA) mitigate these effects, but remain unstable in low sample regimes, which in turn may jeopardize the fairness goal. To address both biases and instability, we propose Fair coVariance Neural Networks (FVNNs), which perform graph convolutions on the covariance matrix for both fair and accurate predictions. Our FVNNs provide a flexible model compatible with several existing bias mitigation techniques. In particular, FVNNs allow for mitigating the bias in two ways: first, they operate on fair covariance estimates that remove biases from their principal components; second, they are trained in an end-to-end fashion via a fairness regularizer in the loss function so that the model parameters are tailored to solve the task directly in a fair manner. We prove that FVNNs are intrinsically fairer than analogous PCA approaches thanks to their stability in low sample regimes. We validate the robustness and fairness of our model on synthetic and real-world data, showcasing the flexibility of FVNNs along with the tradeoff between fair and accurate performance.
♻ ☆ Self-Attention as a Parametric Endofunctor: A Categorical Framework for Transformer Architectures
Self-attention mechanisms have revolutionised deep learning architectures, yet their core mathematical structures remain incompletely understood. In this work, we develop a category-theoretic framework focusing on the linear components of self-attention. Specifically, we show that the query, key, and value maps naturally define a parametric 1-morphism in the 2-category $\mathbf{Para(Vect)}$. On the underlying 1-category $\mathbf{Vect}$, these maps induce an endofunctor whose iterated composition precisely models multi-layer attention. We further prove that stacking multiple self-attention layers corresponds to constructing the free monad on this endofunctor. For positional encodings, we demonstrate that strictly additive embeddings correspond to monoid actions in an affine sense, while standard sinusoidal encodings, though not additive, retain a universal property among injective (faithful) position-preserving maps. We also establish that the linear portions of self-attention exhibit natural equivariance to permutations of input tokens, and show how the "circuits" identified in mechanistic interpretability can be interpreted as compositions of parametric 1-morphisms. This categorical perspective unifies geometric, algebraic, and interpretability-based approaches to transformer analysis, making explicit the underlying structures of attention. We restrict to linear maps throughout, deferring the treatment of nonlinearities such as softmax and layer normalisation, which require more advanced categorical constructions. Our results build on and extend recent work on category-theoretic foundations for deep learning, offering deeper insights into the algebraic structure of attention mechanisms.
♻ ☆ Private Collaborative Edge Inference via Over-the-Air Computation
We consider collaborative inference at the wireless edge, where each client's model is trained independently on its local dataset. Clients are queried in parallel to make an accurate decision collaboratively. In addition to maximizing the inference accuracy, we also want to ensure the privacy of local models. To this end, we leverage the superposition property of the multiple access channel to implement bandwidth-efficient multi-user inference methods. We propose different methods for ensemble and multi-view classification that exploit over-the-air computation (OAC). We show that these schemes perform better than their orthogonal counterparts with statistically significant differences while using fewer resources and providing privacy guarantees. We also provide experimental results verifying the benefits of the proposed OAC approach to multi-user inference, and perform an ablation study to demonstrate the effectiveness of our design choices. We share the source code of the framework publicly on Github to facilitate further research and reproducibility.
comment: 17 pages, 8 figures. This work extends from our preliminary study presented at the 2022 IEEE International Symposium on Information Theory [1]. arXiv admin note: text overlap with arXiv:2202.03129
♻ ☆ One Language, Many Gaps: Evaluating Dialect Fairness and Robustness of Large Language Models in Reasoning Tasks
Language is not monolithic. While benchmarks, including those designed for multiple languages, are often used as proxies to evaluate the performance of Large Language Models (LLMs), they tend to overlook the nuances of within-language variation, and thus fail to model the experience of speakers of non-standard dialects. Focusing on African American Vernacular English (AAVE), we present the first study aimed at objectively assessing the fairness and robustness of LLMs in handling dialects in canonical reasoning tasks, including algorithm, math, logic, and integrated reasoning. We introduce \textbf{ReDial} (\textbf{Re}asoning with \textbf{Dial}ect Queries), a benchmark containing 1.2K+ parallel query pairs in Standardized English and AAVE. We hire AAVE speakers, including experts with computer science backgrounds, to rewrite seven popular benchmarks, such as HumanEval and GSM8K. With ReDial, we evaluate widely used LLMs, including GPT, Claude, Llama, Mistral, and the Phi model families. Our findings reveal that \textbf{almost all of these widely used models show significant brittleness and unfairness to queries in AAVE}. Our work establishes a systematic and objective framework for analyzing LLM bias in dialectal queries. Moreover, it highlights how mainstream LLMs provide unfair service to dialect speakers in reasoning tasks, laying a critical foundation for relevant future research. Code and data can be accessed at https://github.com/fangru-lin/redial_dialect_robustness_fairness.
♻ ☆ Synthesis and Analysis of Data as Probability Measures with Entropy-Regularized Optimal Transport
We consider synthesis and analysis of probability measures using the entropy-regularized Wasserstein-2 cost and its unbiased version, the Sinkhorn divergence. The synthesis problem consists of computing the barycenter, with respect to these costs, of $m$ reference measures given a set of coefficients belonging to the $m$-dimensional simplex. The analysis problem consists of finding the coefficients for the closest barycenter in the Wasserstein-2 distance to a given measure $\mu$. Under the weakest assumptions on the measures thus far in the literature, we compute the derivative of the entropy-regularized Wasserstein-2 cost. We leverage this to establish a characterization of regularized barycenters as solutions to a fixed-point equation for the average of the entropic maps from the barycenter to the reference measures. This characterization yields a finite-dimensional, convex, quadratic program for solving the analysis problem when $\mu$ is a barycenter. It is shown that these coordinates, as well as the value of the barycenter functional, can be estimated from samples with dimension-independent rates of convergence, a hallmark of entropy-regularized optimal transport, and we verify these rates experimentally. We also establish that barycentric coordinates are stable with respect to perturbations in the Wasserstein-2 metric, suggesting a robustness of these coefficients to corruptions. We employ the barycentric coefficients as features for classification of corrupted point cloud data, and show that compared to neural network baselines, our approach is more efficient in small training data regimes.
comment: 58 pages. Code to reproduce experiments: https://github.com/brendanmallery9/Entropic-Barycenters
♻ ☆ KAN KAN Buff Signed Graph Neural Networks?
Graph Representation Learning focuses on creating embeddings for nodes and edges that capture their features and connections. Graph Neural Networks (GNNs) use neural networks to model complex graph relationships. The Kolmogorov-Arnold Neural Network (KAN) has recently emerged as an alternative to the Multi-Layer Perceptron (MLP), offering better accuracy and interpretability with fewer parameters. KANs have been applied to GNN tasks. This paper introduces the integration of KANs into Signed Graph Convolutional Networks (SGCNs). We evaluate KAN-enhanced SGCNs (KASGCN) on signed community detection and link sign prediction tasks to improve embedding quality in signed networks. While the results show some variability, KASGCN performs competitively with or similarly to the standard SGCN in the functions tested. Its effectiveness depends on the specific context, such as the signed graph and parameter settings.
♻ ☆ Evaluation of Artificial Intelligence Methods for Lead Time Prediction in Non-Cycled Areas of Automotive Production
The present study examines the effectiveness of applying Artificial Intelligence methods in an automotive production environment to predict unknown lead times in a non-cycle-controlled production area. Data structures are analyzed to identify contextual features and then preprocessed using one-hot encoding. Methods selection focuses on supervised machine learning techniques. In supervised learning methods, regression and classification methods are evaluated. Continuous regression based on target size distribution is not feasible. Classification methods analysis shows that Ensemble Learning and Support Vector Machines are the most suitable. Preliminary study results indicate that gradient boosting algorithms LightGBM, XGBoost, and CatBoost yield the best results. After further testing and extensive hyperparameter optimization, the final method choice is the LightGBM algorithm. Depending on feature availability and prediction interval granularity, relative prediction accuracies of up to 90% can be achieved. Further tests highlight the importance of periodic retraining of AI models to accurately represent complex production processes using the database. The research demonstrates that AI methods can be effectively applied to highly variable production data, adding business value by providing an additional metric for various control tasks while outperforming current non AI-based systems.
♻ ☆ Set-Based Training for Neural Network Verification
Neural networks are vulnerable to adversarial attacks, i.e., small input perturbations can significantly affect the outputs of a neural network. Therefore, to ensure safety of safety-critical environments, the robustness of a neural network must be formally verified against input perturbations, e.g., from noisy sensors. To improve the robustness of neural networks and thus simplify the formal verification, we present a novel set-based training procedure in which we compute the set of possible outputs given the set of possible inputs and compute for the first time a gradient set, i.e., each possible output has a different gradient. Therefore, we can directly reduce the size of the output enclosure by choosing gradients toward its center. Small output enclosures increase the robustness of a neural network and, at the same time, simplify its formal verification. The latter benefit is due to the fact that a larger size of propagated sets increases the conservatism of most verification methods. Our extensive evaluation demonstrates that set-based training produces robust neural networks with competitive performance, which can be verified using fast (polynomial-time) verification algorithms due to the reduced output set.
♻ ☆ COOL: Efficient and Reliable Chain-Oriented Objective Logic with Neural Networks Feedback Control for Program Synthesis
Program synthesis methods, whether formal or neural-based, lack fine-grained control and flexible modularity, which limits their adaptation to complex software development. These limitations stem from rigid Domain-Specific Language (DSL) frameworks and neural network incorrect predictions. To this end, we propose the Chain of Logic (CoL), which organizes the synthesis process into an activity flow and provides heuristic control to guide the process. Furthermore, by integrating neural networks with libraries and introducing a Neural Network Feedback Control (NNFC) mechanism, our approach modularizes synthesis and mitigates the impact of neural network mispredictions. Experiments on relational and symbolic synthesis tasks show that CoL significantly enhances the efficiency and reliability of DSL program synthesis across multiple metrics. Specifically, CoL improves accuracy by 70% while reducing tree operations by 91% and time by 95%. Additionally, NNFC further boosts accuracy by 6%, with a 64% reduction in tree operations under challenging conditions such as insufficient training data, increased difficulty, and multidomain synthesis. These improvements confirm COOL as a highly efficient and reliable program synthesis framework.
comment: 31 pages, 11 figures
♻ ☆ MoPE: Mixture of Prompt Experts for Parameter-Efficient and Scalable Multimodal Fusion
Despite the demonstrated parameter efficiency of prompt-based multimodal fusion methods, their limited adaptivity and expressiveness often result in suboptimal performance compared to other tuning approaches. In this paper, we introduce the Mixture of Prompt Experts (MoPE), the first technique designed to overcome these limitations by decomposing standard prompts to capture instance-level features adaptively. Building on this decomposition, MoPE enhances prompt fusion's expressiveness by leveraging multimodal pairing priors to route the most effective prompt for each instance dynamically. Compared to vanilla prompting, our MoPE-based fusion method exhibits greater expressiveness, scaling more effectively with the training data and the overall number of trainable parameters. We also investigate regularization terms for expert routing, which lead to emergent expert specialization with enhanced adaptiveness and interpretablity. Extensive experiments across six multimodal datasets spanning four modalities demonstrate state-of-the-art performance for prompt fusion, matching or even surpassing the performance of fine-tuning while requiring only 0.8% of the trainable parameters. Project homepage: https://github.com/songrise/MoPE
comment: Under Review, Extended version of arxiv:2312.03734
♻ ☆ Layer-Adaptive State Pruning for Deep State Space Models NeurIPS 2024
Due to the lack of state dimension optimization methods, deep state space models (SSMs) have sacrificed model capacity, training search space, or stability to alleviate computational costs caused by high state dimensions. In this work, we provide a structured pruning method for SSMs, Layer-Adaptive STate pruning (LAST), which reduces the state dimension of each layer in minimizing model-level output energy loss by extending modal truncation for a single system. LAST scores are evaluated using the $\mathcal{H}_{\infty}$ norms of subsystems and layer-wise energy normalization. The scores serve as global pruning criteria, enabling cross-layer comparison of states and layer-adaptive pruning. Across various sequence benchmarks, LAST optimizes previous SSMs, revealing the redundancy and compressibility of their state spaces. Notably, we demonstrate that, on average, pruning 33% of states still maintains performance with 0.52% accuracy loss in multi-input multi-output SSMs without retraining. Code is available at https://github.com/msgwak/LAST.
comment: NeurIPS 2024
♻ ☆ Data-driven Bayesian State Estimation with Compressed Measurement of Model-free Process using Semi-supervised Learning SP
The research topic is: data-driven Bayesian state estimation with compressed measurement (BSCM) of model-free process, say for a (causal) tracking application. The dimension of the temporal measurement vector is lower than the dimension of the temporal state vector to be estimated. Hence the state estimation problem is an underdetermined inverse problem. The underlying dynamical model of the states is assumed to be unknown and hence, we use the terminology 'model-free process'. In absence of the dynamical model, we can not employ traditional model-driven methods like Kalman Filter (KF) and Particle Filter (PF), and instead require data-driven methods. We first experimentally show that two existing unsupervised learning-based data-driven methods fail to address the BSCM problem for model-free process; they are - data-driven nonlinear state estimation (DANSE) method and deep Markov model (DMM) method. The unsupervised learning uses unlabelled data comprised of only noisy, linear measurements. While DANSE provides a good predictive / forecasting performance to model the temporal measurement data as time-series, its unsupervised learning lacks a regularization for state estimation. We then investigate the use of a semi-supervised learning approach, and develop a semi-supervised learning-based DANSE method, referred to as SemiDANSE. In SemiDANSE, we use a limited amount of labelled data along-with a large amount of unlabelled data, and that helps to bring the desired regularization for addressing the BSCM problem. The labelled data means pairwise measurement-and-state data. Using three chaotic dynamical systems (or processes) with nonlinear dynamical models as benchmark, we show that the data-driven SemiDANSE provides competitive performance for BSCM against a hybrid method called KalmanNet and two model-driven methods -- an extended KF (EKF) and an unscented KF (UKF).
comment: 14 pages, under review at IEEE TSP
♻ ☆ FoMo: A Foundation Model for Mobile Traffic Forecasting with Diffusion Model
Mobile traffic forecasting allows operators to anticipate network dynamics and performance in advance, offering substantial potential for enhancing service quality and improving user experience. However, existing models are often task-oriented and are trained with tailored data, which limits their effectiveness in diverse mobile network tasks of Base Station (BS) deployment, resource allocation, energy optimization, etc. and hinders generalization across different urban environments. Foundation models have made remarkable strides across various domains of NLP and CV due to their multi-tasking adaption and zero/few-shot learning capabilities. In this paper, we propose an innovative Foundation model for Mo}bile traffic forecasting (FoMo), aiming to handle diverse forecasting tasks of short/long-term predictions and distribution generation across multiple cities to support network planning and optimization. FoMo combines diffusion models and transformers, where various spatio-temporal masks are proposed to enable FoMo to learn intrinsic features of different tasks, and a contrastive learning strategy is developed to capture the correlations between mobile traffic and urban contexts, thereby improving its transfer learning capability. Extensive experiments on 9 real-world datasets demonstrate that FoMo outperforms current models concerning diverse forecasting tasks and zero/few-shot learning, showcasing a strong universality.
comment: 11 pages, 7 figures
♻ ☆ Generating Less Certain Adversarial Examples Improves Robust Generalization
This paper revisits the robust overfitting phenomenon of adversarial training. Observing that models with better robust generalization performance are less certain in predicting adversarially generated training inputs, we argue that overconfidence in predicting adversarial examples is a potential cause. Therefore, we hypothesize that generating less certain adversarial examples improves robust generalization, and propose a formal definition of adversarial certainty that captures the variance of the model's predicted logits on adversarial examples. Our theoretical analysis of synthetic distributions characterizes the connection between adversarial certainty and robust generalization. Accordingly, built upon the notion of adversarial certainty, we develop a general method to search for models that can generate training-time adversarial inputs with reduced certainty, while maintaining the model's capability in distinguishing adversarial examples. Extensive experiments on image benchmarks demonstrate that our method effectively learns models with consistently improved robustness and mitigates robust overfitting, confirming the importance of generating less certain adversarial examples for robust generalization. Our implementations are available as open-source code at: https://github.com/TrustMLRG/AdvCertainty.
comment: Published in Transactions on Machine Learning Research (TMLR)
♻ ☆ A User's Guide to $\texttt{KSig}$: GPU-Accelerated Computation of the Signature Kernel
The signature kernel is a positive definite kernel for sequential and temporal data that has become increasingly popular in machine learning applications due to powerful theoretical guarantees, strong empirical performance, and recently introduced various scalable variations. In this chapter, we give a short introduction to $\texttt{KSig}$, a $\texttt{Scikit-Learn}$ compatible Python package that implements various GPU-accelerated algorithms for computing signature kernels, and performing downstream learning tasks. We also introduce a new algorithm based on tensor sketches which gives strong performance compared to existing algorithms. The package is available at https://github.com/tgcsaba/ksig.
♻ ☆ Exploring Gradient Subspaces: Addressing and Overcoming LoRA's Limitations in Federated Fine-Tuning of Large Language Models
Large Language Models (LLMs) have demonstrated remarkable capabilities across various domains, particularly in task generalization for both text and vision data. While fine-tuning these models can significantly enhance their performance on specific downstream tasks, it often requires high-quality data that cannot be shared due to privacy concerns. Federated Learning (FL) offers a promising solution for collaborative training without direct data sharing. However, many parameter-efficient fine-tuning strategies for LLMs in FL, particularly those based on Low-Rank Adaptation (LoRA), face limitations. In this paper, we critically analyze the convergence and performance guarantees of popular FL frameworks utilizing LoRA, highlighting its suboptimal nature due to constrained subspace learning of low-rank matrices. This limitation hinders effective fine-tuning of LLMs in federated settings. Through rigorous analytical and empirical evaluations, we demonstrate that direct weight averaging outperforms LoRA-based strategies, leading to superior performance for fine-tuned models. Our comprehensive comparison unmasks inefficiencies in LoRA approaches and underscores the advantages of direct weight aggregation. We extend our analysis to low-rank gradient-based optimizers, such as GaLore, used during local training steps. Our findings show that GaLore along with direct-weight aggregation is a more effective approach, outperforming federated LoRA methods like FlexLoRA and FFA-LoRA across both text and image modalities. While privacy remains paramount in FL discourse, our focus is on assessing performance outcomes of federated fine-tuned models and evaluating various FL frameworks from both theoretical and empirical perspectives. Our findings advocate reassessing the reliance on LoRA within FL contexts, paving the way for more efficient training methodologies.
♻ ☆ Random Policy Enables In-Context Reinforcement Learning within Trust Horizons
Pretrained foundation models have exhibited extraordinary in-context learning performance, allowing zero-shot generalization to new tasks not encountered during pretraining. In the case of reinforcement learning (RL), in-context RL (ICRL) emerges when pretraining FMs on decision-making problems in an autoregressive-supervised manner. Nevertheless, current state-of-the-art ICRL algorithms, like Algorithm Distillation, Decision Pretrained Transformer and Decision Importance Transformer, impose stringent requirements on the pretraining dataset concerning the source policies, context information, and action labels. Notably, these algorithms either demand optimal policies or require varying degrees of well-trained behavior policies for all pretraining environments. This significantly hinders the application of ICRL to real-world scenarios, where acquiring optimal or well-trained policies for a substantial volume of real-world training environments can be intractable. To overcome this challenge, we introduce a novel approach, termed State-Action Distillation (SAD), that allows to generate an effective pretraining dataset guided solely by random policies. In particular, SAD selects query states and corresponding action labels by distilling outstanding state-action pairs from the entire state and action spaces by using random policies within a trust horizon, and then inherits the classical autoregressive-supervised mechanism during pretraining. To the best of our knowledge, this is the first work that enables effective ICRL under random policies and random contexts. We also establish quantitative analysis of the trustworthiness as well as the performance guarantees of SAD. Moreover, our empirical results across multiple popular ICRL benchmark environments demonstrate that, on average, SAD outperforms the best baseline by 236.3% in the offline evaluation and by 135.2% in the online evaluation.
♻ ☆ Doubly-Bounded Queue for Constrained Online Learning: Keeping Pace with Dynamics of Both Loss and Constraint AAAI 2025
We consider online convex optimization with time-varying constraints and conduct performance analysis using two stringent metrics: dynamic regret with respect to the online solution benchmark, and hard constraint violation that does not allow any compensated violation over time. We propose an efficient algorithm called Constrained Online Learning with Doubly-bounded Queue (COLDQ), which introduces a novel virtual queue that is both lower and upper bounded, allowing tight control of the constraint violation without the need for the Slater condition. We prove via a new Lyapunov drift analysis that COLDQ achieves $O(T^\frac{1+V_x}{2})$ dynamic regret and $O(T^{V_g})$ hard constraint violation, where $V_x$ and $V_g$ capture the dynamics of the loss and constraint functions. For the first time, the two bounds smoothly approach to the best-known $O(T^\frac{1}{2})$ regret and $O(1)$ violation, as the dynamics of the losses and constraints diminish. For strongly convex loss functions, COLDQ matches the best-known $O(\log{T})$ static regret while maintaining the $O(T^{V_g})$ hard constraint violation. We further introduce an expert-tracking variation of COLDQ, which achieves the same performance bounds without any prior knowledge of the system dynamics. Simulation results demonstrate that COLDQ outperforms the state-of-the-art approaches.
comment: To appear in AAAI 2025
♻ ☆ Multi-matrix Factorization Attention
We propose novel attention architectures, Multi-matrix Factorization Attention (MFA) and MFA-Key-Reuse (MFA-KR). Existing variants for standard Multi-Head Attention (MHA), including SOTA methods like MLA, fail to maintain as strong performance under stringent Key-Value cache (KV cache) constraints. MFA enhances model capacity by efficiently scaling up both the number and dimension of attention heads through low-rank matrix factorization in the Query-Key (QK) circuit. Extending MFA, MFA-KR further reduces memory requirements by repurposing the key cache as value through value projection re-parameterization. MFA's design enables strong model capacity when working under tight KV cache budget, while MFA-KR is suitable for even harsher KV cache limits with minor performance trade-off. Notably, in our extensive and large-scale experiments, the proposed architecture outperforms MLA and performs comparably to MHA, while reducing KV cache usage by up to 56% and 93.7%, respectively.
♻ ☆ AdaSociety: An Adaptive Environment with Social Structures for Multi-Agent Decision-Making NeurIPS
Traditional interactive environments limit agents' intelligence growth with fixed tasks. Recently, single-agent environments address this by generating new tasks based on agent actions, enhancing task diversity. We consider the decision-making problem in multi-agent settings, where tasks are further influenced by social connections, affecting rewards and information access. However, existing multi-agent environments lack a combination of adaptive physical surroundings and social connections, hindering the learning of intelligent behaviors. To address this, we introduce AdaSociety, a customizable multi-agent environment featuring expanding state and action spaces, alongside explicit and alterable social structures. As agents progress, the environment adaptively generates new tasks with social structures for agents to undertake. In AdaSociety, we develop three mini-games showcasing distinct social structures and tasks. Initial results demonstrate that specific social structures can promote both individual and collective benefits, though current reinforcement learning and LLM-based algorithms show limited effectiveness in leveraging social structures to enhance performance. Overall, AdaSociety serves as a valuable research platform for exploring intelligence in diverse physical and social settings. The code is available at https://github.com/bigai-ai/AdaSociety.
comment: Accepted at NeurIPS D&B 2024
♻ ☆ Lean Attention: Hardware-Aware Scalable Attention Mechanism for the Decode-Phase of Transformers
Transformer-based models have emerged as one of the most widely used architectures for natural language processing, natural language generation, and image generation. The size of the state-of-the-art models has increased steadily reaching billions of parameters. These huge models are memory hungry and incur significant inference latency even on cutting edge AI-accelerators, such as GPUs. Specifically, the time and memory complexity of the attention operation is quadratic in terms of the total context length, i.e., prompt and output tokens. Thus, several optimizations such as key-value tensor caching and FlashAttention computation have been proposed to deliver the low latency demands of applications relying on such large models. However, these techniques do not cater to the computationally distinct nature of different phases during inference. To that end, we propose LeanAttention, a scalable technique of computing self-attention for the token-generation phase (decode-phase) of decoder-only transformer models. LeanAttention enables scaling the attention mechanism implementation for the challenging case of long context lengths by re-designing the execution flow for the decode-phase. We identify that the associative property of online softmax can be treated as a reduction operation thus allowing us to parallelize the attention computation over these large context lengths. We extend the "stream-K" style reduction of tiled calculation to self-attention to enable parallel computation resulting in an average of 2.6x attention execution speedup over FlashAttention-2 and up to 8.33x speedup for 512k context lengths.
comment: 13 pages, 10 figures
♻ ☆ Poisoning Attacks on Federated Learning-based Wireless Traffic Prediction
Federated Learning (FL) offers a distributed framework to train a global control model across multiple base stations without compromising the privacy of their local network data. This makes it ideal for applications like wireless traffic prediction (WTP), which plays a crucial role in optimizing network resources, enabling proactive traffic flow management, and enhancing the reliability of downstream communication-aided applications, such as IoT devices, autonomous vehicles, and industrial automation systems. Despite its promise, the security aspects of FL-based distributed wireless systems, particularly in regression-based WTP problems, remain inadequately investigated. In this paper, we introduce a novel fake traffic injection (FTI) attack, designed to undermine the FL-based WTP system by injecting fabricated traffic distributions with minimal knowledge. We further propose a defense mechanism, termed global-local inconsistency detection (GLID), which strategically removes abnormal model parameters that deviate beyond a specific percentile range estimated through statistical methods in each dimension. Extensive experimental evaluations, performed on real-world wireless traffic datasets, demonstrate that both our attack and defense strategies significantly outperform existing baselines.
comment: Accepted by IFIP/IEEE Networking 2024
♻ ☆ Radar Signal Recognition through Self-Supervised Learning and Domain Adaptation
Automatic radar signal recognition (RSR) plays a pivotal role in electronic warfare (EW), as accurately classifying radar signals is critical for informing decision-making processes. Recent advances in deep learning have shown significant potential in improving RSR performance in domains with ample annotated data. However, these methods fall short in EW scenarios where annotated RF data are scarce or impractical to obtain. To address these challenges, we introduce a self-supervised learning (SSL) method which utilises masked signal modelling and RF domain adaption to enhance RSR performance in environments with limited RF samples and labels. Specifically, we investigate pre-training masked autoencoders (MAE) on baseband in-phase and quadrature (I/Q) signals from various RF domains and subsequently transfer the learned representation to the radar domain, where annotated data are limited. Empirical results show that our lightweight self-supervised ResNet model with domain adaptation achieves up to a 17.5% improvement in 1-shot classification accuracy when pre-trained on in-domain signals (i.e., radar signals) and up to a 16.31% improvement when pre-trained on out-of-domain signals (i.e., comm signals), compared to its baseline without SSL. We also provide reference results for several MAE designs and pre-training strategies, establishing a new benchmark for few-shot radar signal classification.
comment: 5 pages, 9 figures
♻ ☆ Counterfactually Fair Reinforcement Learning via Sequential Data Preprocessing
When applied in healthcare, reinforcement learning (RL) seeks to dynamically match the right interventions to subjects to maximize population benefit. However, the learned policy may disproportionately allocate efficacious actions to one subpopulation, creating or exacerbating disparities in other socioeconomically-disadvantaged subgroups. These biases tend to occur in multi-stage decision making and can be self-perpetuating, which if unaccounted for could cause serious unintended consequences that limit access to care or treatment benefit. Counterfactual fairness (CF) offers a promising statistical tool grounded in causal inference to formulate and study fairness. In this paper, we propose a general framework for fair sequential decision making. We theoretically characterize the optimal CF policy and prove its stationarity, which greatly simplifies the search for optimal CF policies by leveraging existing RL algorithms. The theory also motivates a sequential data preprocessing algorithm to achieve CF decision making under an additive noise assumption. We prove and then validate our policy learning approach in controlling unfairness and attaining optimal value through simulations. Analysis of a digital health dataset designed to reduce opioid misuse shows that our proposal greatly enhances fair access to counseling.
♻ ☆ Computational and Statistical Asymptotic Analysis of the JKO Scheme for Iterative Algorithms to update distributions
The seminal paper of Jordan, Kinderlehrer, and Otto introduced what is now widely known as the JKO scheme, an iterative algorithmic framework for computing distributions. This scheme can be interpreted as a Wasserstein gradient flow and has been successfully applied in machine learning contexts, such as deriving policy solutions in reinforcement learning. In this paper, we extend the JKO scheme to accommodate models with unknown parameters. Specifically, we develop statistical methods to estimate these parameters and adapt the JKO scheme to incorporate the estimated values. To analyze the adopted statistical JKO scheme, we establish an asymptotic theory via stochastic partial differential equations that describes its limiting dynamic behavior. Our framework allows both the sample size used in parameter estimation and the number of algorithmic iterations to go to infinity. This study offers a unified framework for joint computational and statistical asymptotic analysis of the statistical JKO scheme. On the computational side, we examine the scheme's dynamic behavior as the number of iterations increases, while on the statistical side, we investigate the large-sample behavior of the resulting distributions computed through the scheme. We conduct numerical simulations to evaluate the finite-sample performance of the proposed methods and validate the developed asymptotic theory.
♻ ☆ ELDER: Enhancing Lifelong Model Editing with Mixture-of-LoRA AAAI-25
Large language models (LLMs) require model editing to efficiently update specific knowledge within them and avoid factual errors. Most model editing methods are solely designed for single-time use and result in a significant forgetting effect in lifelong editing scenarios, where sequential edits are conducted over time. Previous approaches manage sequential edits by freezing original parameters and discretely allocating new parameters for each knowledge update. However, these methods lack robustness to minor input variations due to the discrete mapping between data and parameters. To overcome this challenge, we propose ELDER, a novel approach to create a continuous association between data and adapters. ELDER integrates multiple LoRAs through a router network and is trained to establish a smooth data-adapter association, thereby enhancing the edit robustness and generalization of semantically equivalent inputs. To ensure inputs containing the same knowledge will be processed by the same LoRAs, we design a novel loss to guide the model link LoRA allocations with edit knowledge. Furthermore, we propose a deferral mechanism to retain the original LLM capabilities post-edit. Extensive experiments on GPT-2 XL and LLaMA2-7B demonstrate that ELDER effectively edits models in the lifelong setting, outperforming eight baselines while exhibiting strong scalability and preserving LLMs' general abilities on downstream tasks. Our code is available at https://github.com/JiaangL/ELDER.
comment: Accepted by AAAI-25
♻ ☆ Retrieval-Reasoning Large Language Model-based Synthetic Clinical Trial Generation
Machine learning (ML) exhibits promise in the clinical domain. However, it is constrained by data scarcity and ethical considerations, as the generation of clinical trials presents significant challenges due to stringent privacy regulations, high costs, and the extended duration required for conducting studies with human participants. Despite the advancements of large language models (LLMs) in general generation tasks, their potential in facilitating the generation of synthetic clinical trials is under-explored. To address this gap, we introduce a novel Retrieval-Reasoning few-shot framework that leverages LLMs to generate artificial yet realistic and diverse clinical trials with binary success/failure labels. Experiments conducted on real clinical trials from the \url{ClinicalTrials.gov} database demonstrate that our synthetic data can effectively augment real datasets. Furthermore, by fine-tuning a pre-trained model as a binary classifier on synthetic clinical trial datasets, we demonstrate that this augmentation enhances model training for downstream tasks such as trial outcome prediction. Our findings suggest that LLMs for synthetic clinical trial generation hold promise for accelerating clinical research and upholding ethical standards for patient privacy. The code is publicly available at https://anonymous.4open.science/r/Retrieval_Reasoning_Clinical_Trial_Generation-3EC4.
♻ ☆ AI Foundation Models for Wearable Movement Data in Mental Health Research
Pretrained foundation models and transformer architectures have driven the success of large language models (LLMs) and other modern AI breakthroughs. However, similar advancements in health data modeling remain limited due to the need for innovative adaptations. Wearable movement data offers a valuable avenue for exploration, as it's a core feature in nearly all commercial smartwatches, well established in clinical and mental health research, and the sequential nature of the data shares similarities to language. We introduce the Pretrained Actigraphy Transformer (PAT), the first open source foundation model designed for time-series wearable movement data. Leveraging transformer-based architectures and novel techniques, such as patch embeddings, and pretraining on data from 29,307 participants in a national U.S. sample, PAT achieves state-of-the-art performance in several mental health prediction tasks. PAT is also lightweight and easily interpretable, making it a robust tool for mental health research. GitHub: https://github.com/njacobsonlab/Pretrained-Actigraphy-Transformer/
♻ ☆ Energy-Efficient Split Learning for Fine-Tuning Large Language Models in Edge Networks
In this letter, we propose an energy-efficient split learning (SL) framework for fine-tuning large language models (LLMs) using geo-distributed personal data at the network edge, where LLMs are split and alternately across massive mobile devices and an edge server. Considering the device heterogeneity and channel dynamics in edge networks, a \underline{C}ut l\underline{A}yer and computing \underline{R}esource \underline{D}ecision (CARD) algorithm is developed to minimize training delay and energy consumption. Simulation results demonstrate that the proposed approach reduces the average training delay and server's energy consumption by 70.8% and 53.1%, compared to the benchmarks, respectively.
comment: 5 pages, 6 figures
♻ ☆ Can Go AIs be adversarially robust? AAAI 2025
Prior work found that superhuman Go AIs can be defeated by simple adversarial strategies, especially "cyclic" attacks. In this paper, we study whether adding natural countermeasures can achieve robustness in Go, a favorable domain for robustness since it benefits from incredible average-case capability and a narrow, innately adversarial setting. We test three defenses: adversarial training on hand-constructed positions, iterated adversarial training, and changing the network architecture. We find that though some of these defenses protect against previously discovered attacks, none withstand freshly trained adversaries. Furthermore, most of the reliably effective attacks these adversaries discover are different realizations of the same overall class of cyclic attacks. Our results suggest that building robust AI systems is challenging even with extremely superhuman systems in some of the most tractable settings, and highlight two key gaps: efficient generalization of defenses, and diversity in training. For interactive examples of attacks and a link to our codebase, see https://goattack.far.ai.
comment: 63 pages, AAAI 2025
♻ ☆ Physically Guided Deep Unsupervised Inversion for 1D Magnetotelluric Models
The global demand for unconventional energy sources such as geothermal energy and white hydrogen requires new exploration techniques for precise subsurface structure characterization and potential reservoir identification. The Magnetotelluric (MT) method is crucial for these tasks, providing critical information on the distribution of subsurface electrical resistivity at depths ranging from hundreds to thousands of meters. However, traditional iterative algorithm-based inversion methods require the adjustment of multiple parameters, demanding time-consuming and exhaustive tuning processes to achieve proper cost function minimization. Recent advances have incorporated deep learning algorithms for MT inversion, primarily based on supervised learning, and large labeled datasets are needed for training. This work utilizes TensorFlow operations to create a differentiable forward MT operator, leveraging its automatic differentiation capability. Moreover, instead of solving for the subsurface model directly, as classical algorithms perform, this paper presents a new deep unsupervised inversion algorithm guided by physics to estimate 1D MT models. Instead of using datasets with the observed data and their respective model as labels during training, our method employs a differentiable modeling operator that physically guides the cost function minimization, making the proposed method solely dependent on observed data. Therefore, the optimization algorithm updates the network weights to minimize the data misfit. We test the proposed method with field and synthetic data at different acquisition frequencies, demonstrating that the resistivity models obtained are more accurate than those calculated using other techniques.
comment: 5 pages, 6 figures, github repository, submitted to IEEE-GRSL
♻ ☆ $\text{Transformer}^2$: Self-adaptive LLMs
Self-adaptive large language models (LLMs) aim to solve the challenges posed by traditional fine-tuning methods, which are often computationally intensive and static in their ability to handle diverse tasks. We introduce $\text{Transformer}^2$, a novel self-adaptation framework that adapts LLMs for unseen tasks in real-time by selectively adjusting only the singular components of their weight matrices. During inference, $\text{Transformer}^2$ employs a two-pass mechanism: first, a dispatch system identifies the task properties, and then task-specific "expert" vectors, trained using reinforcement learning, are dynamically mixed to obtain targeted behavior for the incoming prompt. Our method outperforms ubiquitous approaches such as LoRA, with fewer parameters and greater efficiency. $\text{Transformer}^2$ demonstrates versatility across different LLM architectures and modalities, including vision-language tasks. $\text{Transformer}^2$ represents a significant leap forward, offering a scalable, efficient solution for enhancing the adaptability and task-specific performance of LLMs, paving the way for truly dynamic, self-organizing AI systems.
comment: 18 panges, 11 figures, 9 tables
♻ ☆ E2ESlack: An End-to-End Graph-Based Framework for Pre-Routing Slack Prediction
Pre-routing slack prediction remains a critical area of research in Electronic Design Automation (EDA). Despite numerous machine learning-based approaches targeting this task, there is still a lack of a truly end-to-end framework that engineers can use to obtain TNS/WNS metrics from raw circuit data at the placement stage. Existing works have demonstrated effectiveness in Arrival Time (AT) prediction but lack a mechanism for Required Arrival Time (RAT) prediction, which is essential for slack prediction and obtaining TNS/WNS metrics. In this work, we propose E2ESlack, an end-to-end graph-based framework for pre-routing slack prediction. The framework includes a TimingParser that supports DEF, SDF and LIB files for feature extraction and graph construction, an arrival time prediction model and a fast RAT estimation module. To the best of our knowledge, this is the first work capable of predicting path-level slacks at the pre-routing stage. We perform extensive experiments and demonstrate that our proposed RAT estimation method outperforms the SOTA ML-based prediction method and also pre-routing STA tool. Additionally, the proposed E2ESlack framework achieves TNS/WNS values comparable to post-routing STA results while saving up to 23x runtime.
♻ ☆ Gradient descent with generalized Newton's method
We propose the generalized Newton's method (GeN) -- a Hessian-informed approach that applies to any optimizer such as SGD and Adam, and covers the Newton-Raphson method as a sub-case. Our method automatically and dynamically selects the learning rate that accelerates the convergence, without the intensive tuning of the learning rate scheduler. In practice, our method is easily implementable, since it only requires additional forward passes with almost zero computational overhead (in terms of training time and memory cost), if the overhead is amortized over many iterations. We present extensive experiments on language and vision tasks (e.g. GPT and ResNet) to showcase that GeN optimizers match the state-of-the-art performance, which was achieved with carefully tuned learning rate schedulers.
♻ ☆ Can AI Help with Your Personal Finances?
In recent years, Large Language Models (LLMs) have emerged as a transformative development in artificial intelligence (AI), drawing significant attention from industry and academia. Trained on vast datasets, these sophisticated AI systems exhibit impressive natural language processing and content generation capabilities. This paper explores the potential of LLMs to address key challenges in personal finance, focusing on the United States. We evaluate several leading LLMs, including OpenAI's ChatGPT, Google's Gemini, Anthropic's Claude, and Meta's Llama, to assess their effectiveness in providing accurate financial advice on topics such as mortgages, taxes, loans, and investments. Our findings show that while these models achieve an average accuracy rate of approximately 70%, they also display notable limitations in certain areas. Specifically, LLMs struggle to provide accurate responses for complex financial queries, with performance varying significantly across different topics. Despite these limitations, the analysis reveals notable improvements in newer versions of these models, highlighting their growing utility for individuals and financial advisors. As these AI systems continue to evolve, their potential for advancing AI-driven applications in personal finance becomes increasingly promising.
♻ ☆ Smartphone-based Eye Tracking System using Edge Intelligence and Model Optimisation
A significant limitation of current smartphone-based eye-tracking algorithms is their low accuracy when applied to video-type visual stimuli, as they are typically trained on static images. Also, the increasing demand for real-time interactive applications like games, VR, and AR on smartphones requires overcoming the limitations posed by resource constraints such as limited computational power, battery life, and network bandwidth. Therefore, we developed two new smartphone eye-tracking techniques for video-type visuals by combining Convolutional Neural Networks (CNN) with two different Recurrent Neural Networks (RNN), namely Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU). Our CNN+LSTM and CNN+GRU models achieved an average Root Mean Square Error of 0.955 cm and 1.091 cm, respectively. To address the computational constraints of smartphones, we developed an edge intelligence architecture to enhance the performance of smartphone-based eye tracking. We applied various optimisation methods like quantisation and pruning to deep learning models for better energy, CPU, and memory usage on edge devices, focusing on real-time processing. Using model quantisation, the model inference time in the CNN+LSTM and CNN+GRU models was reduced by 21.72% and 19.50%, respectively, on edge devices.
comment: I have included the three papers as reference, which are closely related. We have expanded the future work section to provide a more thorough discussion of the concepts of "varying lighting conditions" and "dynamic user environments." We have added a note below Table 4 to clarify the abbreviations' meaning. Elaborated the role of the Domain Expert within the presentation layer in Section 4.1
♻ ☆ ACPO: AI-Enabled Compiler Framework
The key to performance optimization of a program is to decide correctly when a certain transformation should be applied by a compiler. This is an ideal opportunity to apply machine-learning models to speed up the tuning process; while this realization has been around since the late 90s, only recent advancements in ML enabled a practical application of ML to compilers as an end-to-end framework. This paper presents ACPO: An AI-Enabled Compiler Framework, a novel framework that provides LLVM with simple and comprehensive tools to benefit from employing ML models for different optimization passes. We first showcase the high-level view, class hierarchy, and functionalities of ACPO and subsequently, demonstrate \taco{a couple of use cases of ACPO by ML-enabling the Loop Unroll and Function Inlining passes used in LLVM's O3. and finally, describe how ACPO can be leveraged to optimize other passes. Experimental results reveal that the ACPO model for Loop Unroll can gain on average 4%, 3%, 5.4%, and 0.2% compared to LLVM's vanilla O3 optimization when deployed on Polybench, Coral-2, CoreMark, and Graph-500, respectively. Furthermore, by including both Function Inlining and Loop Unroll models, ACPO can provide a combined speedup of 4.5% on Polybench and 2.4% on Cbench when compared with LLVM's O3, respectively.
comment: ACPO (12 pages)
♻ ☆ EPIC: Effective Prompting for Imbalanced-Class Data Synthesis in Tabular Data Classification via Large Language Models NeurIPS 2024
Large language models (LLMs) have demonstrated remarkable in-context learning capabilities across diverse applications. In this work, we explore the effectiveness of LLMs for generating realistic synthetic tabular data, identifying key prompt design elements to optimize performance. We introduce EPIC, a novel approach that leverages balanced, grouped data samples and consistent formatting with unique variable mapping to guide LLMs in generating accurate synthetic data across all classes, even for imbalanced datasets. Evaluations on real-world datasets show that EPIC achieves state-of-the-art machine learning classification performance, significantly improving generation efficiency. These findings highlight the effectiveness of EPIC for synthetic tabular data generation, particularly in addressing class imbalance. Our source code for our work is available at: https://seharanul17.github.io/project-synthetic-tabular-llm/
comment: NeurIPS 2024
♻ ☆ A systematic review of the use of Deep Learning in Satellite Imagery for Agriculture
Agricultural research is essential for increasing food production to meet the requirements of an increasing population in the coming decades. Recently, satellite technology has been improving rapidly and deep learning has seen much success in generic computer vision tasks and many application areas which presents an important opportunity to improve analysis of agricultural land. Here we present a systematic review of 150 studies to find the current uses of deep learning on satellite imagery for agricultural research. Although we identify 5 categories of agricultural monitoring tasks, the majority of the research interest is in crop segmentation and yield prediction. We found that, when used, modern deep learning methods consistently outperformed traditional machine learning across most tasks; the only exception was that Long Short-Term Memory (LSTM) Recurrent Neural Networks did not consistently outperform Random Forests (RF) for yield prediction. The reviewed studies have largely adopted methodologies from generic computer vision, except for one major omission: benchmark datasets are not utilised to evaluate models across studies, making it difficult to compare results. Additionally, some studies have specifically utilised the extra spectral resolution available in satellite imagery, but other divergent properties of satellite images - such as the hugely different scales of spatial patterns - are not being taken advantage of in the reviewed studies.
comment: 23 pages, 5 figures and 10 tables in main paper. Final version, as submitted and accepted at JSTARS
♻ ☆ Double Equivariance for Inductive Link Prediction for Both New Nodes and New Relation Types
The task of fully inductive link prediction in knowledge graphs has gained significant attention, with various graph neural networks being proposed to address it. This task presents greater challenges than traditional inductive link prediction tasks with only new nodes, as models must be capable of zero-shot generalization to both unseen nodes and unseen relation types in the inference graph. Despite the development of novel models, a unifying theoretical understanding of their success remains elusive, and the limitations of these methods are not well-studied. In this work, we introduce the concept of double permutation-equivariant representations and demonstrate its necessity for effective performance in this task. We show that many existing models, despite their diverse architectural designs, conform to this framework. However, we also identify inherent limitations in double permutation-equivariant representations, which restrict these models's ability to learn effectively on datasets with varying characteristics. Our findings suggest that while double equivariance is necessary for meta-learning across knowledge graphs from different domains, it is not sufficient. There remains a fundamental gap between double permutation-equivariant models and the concept of foundation models designed to learn patterns across all domains.
♻ ☆ High-dimensional learning of narrow neural networks
Recent years have been marked with the fast-pace diversification and increasing ubiquity of machine learning applications. Yet, a firm theoretical understanding of the surprising efficiency of neural networks to learn from high-dimensional data still proves largely elusive. In this endeavour, analyses inspired by statistical physics have proven instrumental, enabling the tight asymptotic characterization of the learning of neural networks in high dimensions, for a broad class of solvable models. This manuscript reviews the tools and ideas underlying recent progress in this line of work. We introduce a generic model -- the sequence multi-index model -- which encompasses numerous previously studied models as special instances. This unified framework covers a broad class of machine learning architectures with a finite number of hidden units, including multi-layer perceptrons, autoencoders, attention mechanisms; and tasks, including (un)supervised learning, denoising, contrastive learning, in the limit of large data dimension, and comparably large number of samples. We explicate in full detail the analysis of the learning of sequence multi-index models, using statistical physics techniques such as the replica method and approximate message-passing algorithms. This manuscript thus provides a unified presentation of analyses reported in several previous works, and a detailed overview of central techniques in the field of statistical physics of machine learning. This review should be a useful primer for machine learning theoreticians curious of statistical physics approaches; it should also be of value to statistical physicists interested in the transfer of such ideas to the study of neural networks.
♻ ☆ Expressive Text-to-Image Generation with Rich Text
Plain text has become a prevalent interface for text-to-image synthesis. However, its limited customization options hinder users from accurately describing desired outputs. For example, plain text makes it hard to specify continuous quantities, such as the precise RGB color value or importance of each word. Furthermore, creating detailed text prompts for complex scenes is tedious for humans to write and challenging for text encoders to interpret. To address these challenges, we propose using a rich-text editor supporting formats such as font style, size, color, and footnote. We extract each word's attributes from rich text to enable local style control, explicit token reweighting, precise color rendering, and detailed region synthesis. We achieve these capabilities through a region-based diffusion process. We first obtain each word's region based on attention maps of a diffusion process using plain text. For each region, we enforce its text attributes by creating region-specific detailed prompts and applying region-specific guidance, and maintain its fidelity against plain-text generation through region-based injections. We present various examples of image generation from rich text and demonstrate that our method outperforms strong baselines with quantitative evaluations.
comment: Project webpage: https://rich-text-to-image.github.io/
♻ ☆ Neural Network Emulator for Atmospheric Chemical ODE
Modeling atmospheric chemistry is complex and computationally intense. Given the recent success of Deep neural networks in digital signal processing, we propose a Neural Network Emulator for fast chemical concentration modeling. We consider atmospheric chemistry as a time-dependent Ordinary Differential Equation. To extract the hidden correlations between initial states and future time evolution, we propose ChemNNE, an Attention based Neural Network Emulator (NNE) that can model the atmospheric chemistry as a neural ODE process. To efficiently simulate the chemical changes, we propose the sinusoidal time embedding to estimate the oscillating tendency over time. More importantly, we use the Fourier neural operator to model the ODE process for efficient computation. We also propose three physical-informed losses to supervise the training optimization. To evaluate our model, we propose a large-scale chemical dataset that can be used for neural network training and evaluation. The extensive experiments show that our approach achieves state-of-the-art performance in modeling accuracy and computational speed.
comment: 25 pages, 8 figures
♻ ☆ MassSpecGym: A benchmark for the discovery and identification of molecules
The discovery and identification of molecules in biological and environmental samples is crucial for advancing biomedical and chemical sciences. Tandem mass spectrometry (MS/MS) is the leading technique for high-throughput elucidation of molecular structures. However, decoding a molecular structure from its mass spectrum is exceptionally challenging, even when performed by human experts. As a result, the vast majority of acquired MS/MS spectra remain uninterpreted, thereby limiting our understanding of the underlying (bio)chemical processes. Despite decades of progress in machine learning applications for predicting molecular structures from MS/MS spectra, the development of new methods is severely hindered by the lack of standard datasets and evaluation protocols. To address this problem, we propose MassSpecGym -- the first comprehensive benchmark for the discovery and identification of molecules from MS/MS data. Our benchmark comprises the largest publicly available collection of high-quality labeled MS/MS spectra and defines three MS/MS annotation challenges: \textit{de novo} molecular structure generation, molecule retrieval, and spectrum simulation. It includes new evaluation metrics and a generalization-demanding data split, therefore standardizing the MS/MS annotation tasks and rendering the problem accessible to the broad machine learning community. MassSpecGym is publicly available at \url{https://github.com/pluskal-lab/MassSpecGym}.
♻ ☆ A Parameter-Efficient Quantum Anomaly Detection Method on a Superconducting Quantum Processor
Quantum machine learning has gained attention for its potential to address computational challenges. However, whether those algorithms can effectively solve practical problems and outperform their classical counterparts, especially on current quantum hardware, remains a critical question. In this work, we propose a novel quantum machine learning method, called Quantum Support Vector Data Description (QSVDD), for practical image anomaly detection, which aims to achieve both parameter efficiency and superior accuracy compared to classical models. Emulation results indicate that QSVDD demonstrates favourable recognition capabilities compared to classical baselines, achieving an average accuracy of over 90% on benchmarks with significantly fewer trainable parameters. Theoretical analysis confirms that QSVDD has a comparable expressivity to classical counterparts while requiring only a fraction of the parameters. Furthermore, we demonstrate the first implementation of a quantum anomaly detection method for general image datasets on a superconducting quantum processor. Specifically, we achieve an accuracy of over 80% with only 16 parameters on the device, providing initial evidence of QSVDD's practical viability in the noisy intermediate-scale quantum era and highlighting its significant reduction in parameter requirements.
comment: 30 pages, 13 figures
Artificial Intelligence 178
☆ PokerBench: Training Large Language Models to become Professional Poker Players AAAI 2025
We introduce PokerBench - a benchmark for evaluating the poker-playing abilities of large language models (LLMs). As LLMs excel in traditional NLP tasks, their application to complex, strategic games like poker poses a new challenge. Poker, an incomplete information game, demands a multitude of skills such as mathematics, reasoning, planning, strategy, and a deep understanding of game theory and human psychology. This makes Poker the ideal next frontier for large language models. PokerBench consists of a comprehensive compilation of 11,000 most important scenarios, split between pre-flop and post-flop play, developed in collaboration with trained poker players. We evaluate prominent models including GPT-4, ChatGPT 3.5, and various Llama and Gemma series models, finding that all state-of-the-art LLMs underperform in playing optimal poker. However, after fine-tuning, these models show marked improvements. We validate PokerBench by having models with different scores compete with each other, demonstrating that higher scores on PokerBench lead to higher win rates in actual poker games. Through gameplay between our fine-tuned model and GPT-4, we also identify limitations of simple supervised fine-tuning for learning optimal playing strategy, suggesting the need for more advanced methodologies for effectively training language models to excel in games. PokerBench thus presents a unique benchmark for a quick and reliable evaluation of the poker-playing ability of LLMs as well as a comprehensive benchmark to study the progress of LLMs in complex game-playing scenarios. The dataset and code will be made available at: \url{https://github.com/pokerllm/pokerbench}.
comment: AAAI 2025
☆ ADAM-1: AI and Bioinformatics for Alzheimer's Detection and Microbiome-Clinical Data Integrations
The Alzheimer's Disease Analysis Model Generation 1 (ADAM) is a multi-agent large language model (LLM) framework designed to integrate and analyze multi-modal data, including microbiome profiles, clinical datasets, and external knowledge bases, to enhance the understanding and detection of Alzheimer's disease (AD). By leveraging retrieval-augmented generation (RAG) techniques along with its multi-agent architecture, ADAM-1 synthesizes insights from diverse data sources and contextualizes findings using literature-driven evidence. Comparative evaluation against XGBoost revealed similar mean F1 scores but significantly reduced variance for ADAM-1, highlighting its robustness and consistency, particularly in small laboratory datasets. While currently tailored for binary classification tasks, future iterations aim to incorporate additional data modalities, such as neuroimaging and biomarkers, to broaden the scalability and applicability for Alzheimer's research and diagnostics.
comment: 16 pages, 16 figures
☆ Diffusion Adversarial Post-Training for One-Step Video Generation
The diffusion models are widely used for image and video generation, but their iterative generation process is slow and expansive. While existing distillation approaches have demonstrated the potential for one-step generation in the image domain, they still suffer from significant quality degradation. In this work, we propose Adversarial Post-Training (APT) against real data following diffusion pre-training for one-step video generation. To improve the training stability and quality, we introduce several improvements to the model architecture and training procedures, along with an approximated R1 regularization objective. Empirically, our experiments show that our adversarial post-trained model, Seaweed-APT, can generate 2-second, 1280x720, 24fps videos in real time using a single forward evaluation step. Additionally, our model is capable of generating 1024px images in a single step, achieving quality comparable to state-of-the-art methods.
☆ Polynomial Threshold Functions of Bounded Tree-Width: Some Explainability and Complexity Aspects
The tree-width of a multivariate polynomial is the tree-width of the hypergraph with hyperedges corresponding to its terms. Multivariate polynomials of bounded tree-width have been studied by Makowsky and Meer as a new sparsity condition that allows for polynomial solvability of problems which are intractable in general. We consider a variation on this theme for Boolean variables. A representation of a Boolean function as the sign of a polynomial is called a polynomial threshold representation. We discuss Boolean functions representable as polynomial threshold functions of bounded tree-width and present two applications to Bayesian network classifiers, a probabilistic graphical model. Both applications are in Explainable Artificial Intelligence (XAI), the research area dealing with the black-box nature of many recent machine learning models. We also give a separation result between the representational power of positive and general polynomial threshold functions.
comment: 22 pages, 3 figures. To be published in Festschrift in honor of Johann A. Makowsky
☆ HALoGEN: Fantastic LLM Hallucinations and Where to Find Them
Despite their impressive ability to generate high-quality and fluent text, generative large language models (LLMs) also produce hallucinations: statements that are misaligned with established world knowledge or provided input context. However, measuring hallucination can be challenging, as having humans verify model generations on-the-fly is both expensive and time-consuming. In this work, we release HALoGEN, a comprehensive hallucination benchmark consisting of: (1) 10,923 prompts for generative models spanning nine domains including programming, scientific attribution, and summarization, and (2) automatic high-precision verifiers for each use case that decompose LLM generations into atomic units, and verify each unit against a high-quality knowledge source. We use this framework to evaluate ~150,000 generations from 14 language models, finding that even the best-performing models are riddled with hallucinations (sometimes up to 86% of generated atomic facts depending on the domain). We further define a novel error classification for LLM hallucinations based on whether they likely stem from incorrect recollection of training data (Type A errors), or incorrect knowledge in training data (Type B errors), or are fabrication (Type C errors). We hope our framework provides a foundation to enable the principled study of why generative models hallucinate, and advances the development of trustworthy large language models.
comment: Preprint
☆ Comparative Analysis of Efficient Adapter-Based Fine-Tuning of State-of-the-Art Transformer Models
In this work, we investigate the efficacy of various adapter architectures on supervised binary classification tasks from the SuperGLUE benchmark as well as a supervised multi-class news category classification task from Kaggle. Specifically, we compare classification performance and time complexity of three transformer models, namely DistilBERT, ELECTRA, and BART, using conventional fine-tuning as well as nine state-of-the-art (SoTA) adapter architectures. Our analysis reveals performance differences across adapter architectures, highlighting their ability to achieve comparable or better performance relative to fine-tuning at a fraction of the training time. Similar results are observed on the new classification task, further supporting our findings and demonstrating adapters as efficient and flexible alternatives to fine-tuning. This study provides valuable insights and guidelines for selecting and implementing adapters in diverse natural language processing (NLP) applications.
☆ AI Driven Water Segmentation with deep learning models for Enhanced Flood Monitoring
Flooding is a major natural hazard causing significant fatalities and economic losses annually, with increasing frequency due to climate change. Rapid and accurate flood detection and monitoring are crucial for mitigating these impacts. This study compares the performance of three deep learning models UNet, ResNet, and DeepLabv3 for pixelwise water segmentation to aid in flood detection, utilizing images from drones, in field observations, and social media. This study involves creating a new dataset that augments wellknown benchmark datasets with flood-specific images, enhancing the robustness of the models. The UNet, ResNet, and DeepLab v3 architectures are tested to determine their effectiveness in various environmental conditions and geographical locations, and the strengths and limitations of each model are also discussed here, providing insights into their applicability in different scenarios by predicting image segmentation masks. This fully automated approach allows these models to isolate flooded areas in images, significantly reducing processing time compared to traditional semi-automated methods. The outcome of this study is to predict segmented masks for each image effected by a flood disaster and the validation accuracy of these models. This methodology facilitates timely and continuous flood monitoring, providing vital data for emergency response teams to reduce loss of life and economic damages. It offers a significant reduction in the time required to generate flood maps, cutting down the manual processing time. Additionally, we present avenues for future research, including the integration of multimodal data sources and the development of robust deep learning architectures tailored specifically for flood detection tasks. Overall, our work contributes to the advancement of flood management strategies through innovative use of deep learning technologies.
comment: 8 pages, 6 figures
☆ Eliciting In-context Retrieval and Reasoning for Long-context Large Language Models
Recent advancements in long-context language models (LCLMs) promise to transform Retrieval-Augmented Generation (RAG) by simplifying pipelines. With their expanded context windows, LCLMs can process entire knowledge bases and perform retrieval and reasoning directly -- a capability we define as In-Context Retrieval and Reasoning (ICR^2). However, existing benchmarks like LOFT often overestimate LCLM performance by providing overly simplified contexts. To address this, we introduce ICR^2, a benchmark that evaluates LCLMs in more realistic scenarios by including confounding passages retrieved with strong retrievers. We then propose three methods to enhance LCLM performance: (1) retrieve-then-generate fine-tuning, (2) retrieval-attention-probing, which uses attention heads to filter and de-noise long contexts during decoding, and (3) joint retrieval head training alongside the generation head. Our evaluation of five well-known LCLMs on LOFT and ICR^2 demonstrates significant gains with our best approach applied to Mistral-7B: +17 and +15 points by Exact Match on LOFT, and +13 and +2 points on ICR^2, compared to vanilla RAG and supervised fine-tuning, respectively. It even outperforms GPT-4-Turbo on most tasks despite being a much smaller model.
☆ Engineering LLM Powered Multi-agent Framework for Autonomous CloudOps ICSE 2025
Cloud Operations (CloudOps) is a rapidly growing field focused on the automated management and optimization of cloud infrastructure which is essential for organizations navigating increasingly complex cloud environments. MontyCloud Inc. is one of the major companies in the CloudOps domain that leverages autonomous bots to manage cloud compliance, security, and continuous operations. To make the platform more accessible and effective to the customers, we leveraged the use of GenAI. Developing a GenAI-based solution for autonomous CloudOps for the existing MontyCloud system presented us with various challenges such as i) diverse data sources; ii) orchestration of multiple processes; and iii) handling complex workflows to automate routine tasks. To this end, we developed MOYA, a multi-agent framework that leverages GenAI and balances autonomy with the necessary human control. This framework integrates various internal and external systems and is optimized for factors like task orchestration, security, and error mitigation while producing accurate, reliable, and relevant insights by utilizing Retrieval Augmented Generation (RAG). Evaluations of our multi-agent system with the help of practitioners as well as using automated checks demonstrate enhanced accuracy, responsiveness, and effectiveness over non-agentic approaches across complex workflows.
comment: The paper has been accepted as full paper to CAIN 2025 (https://conf.researchr.org/home/cain-2025), co-located with ICSE 2025 (https://conf.researchr.org/home/icse-2025). The paper was submitted to CAIN for review on 9 November 2024
☆ A Feature-Level Ensemble Model for COVID-19 Identification in CXR Images using Choquet Integral and Differential Evolution Optimization
The COVID-19 pandemic has profoundly impacted billions globally. It challenges public health and healthcare systems due to its rapid spread and severe respiratory effects. An effective strategy to mitigate the COVID-19 pandemic involves integrating testing to identify infected individuals. While RT-PCR is considered the gold standard for diagnosing COVID-19, it has some limitations such as the risk of false negatives. To address this problem, this paper introduces a novel Deep Learning Diagnosis System that integrates pre-trained Deep Convolutional Neural Networks (DCNNs) within an ensemble learning framework to achieve precise identification of COVID-19 cases from Chest X-ray (CXR) images. We combine feature vectors from the final hidden layers of pre-trained DCNNs using the Choquet integral to capture interactions between different DCNNs that a linear approach cannot. We employed Sugeno-$\lambda$ measure theory to derive fuzzy measures for subsets of networks to enable aggregation. We utilized Differential Evolution to estimate fuzzy densities. We developed a TensorFlow-based layer for Choquet operation to facilitate efficient aggregation, due to the intricacies involved in aggregating feature vectors. Experimental results on the COVIDx dataset show that our ensemble model achieved 98\% accuracy in three-class classification and 99.50\% in binary classification, outperforming its components-DenseNet-201 (97\% for three-class, 98.75\% for binary), Inception-v3 (96.25\% for three-class, 98.50\% for binary), and Xception (94.50\% for three-class, 98\% for binary)-and surpassing many previous methods.
☆ Dynamic Pricing in High-Speed Railways Using Multi-Agent Reinforcement Learning
This paper addresses a critical challenge in the high-speed passenger railway industry: designing effective dynamic pricing strategies in the context of competing and cooperating operators. To address this, a multi-agent reinforcement learning (MARL) framework based on a non-zero-sum Markov game is proposed, incorporating random utility models to capture passenger decision making. Unlike prior studies in areas such as energy, airlines, and mobile networks, dynamic pricing for railway systems using deep reinforcement learning has received limited attention. A key contribution of this paper is a parametrisable and versatile reinforcement learning simulator designed to model a variety of railway network configurations and demand patterns while enabling realistic, microscopic modelling of user behaviour, called RailPricing-RL. This environment supports the proposed MARL framework, which models heterogeneous agents competing to maximise individual profits while fostering cooperative behaviour to synchronise connecting services. Experimental results validate the framework, demonstrating how user preferences affect MARL performance and how pricing policies influence passenger choices, utility, and overall system dynamics. This study provides a foundation for advancing dynamic pricing strategies in railway systems, aligning profitability with system-wide efficiency, and supporting future research on optimising pricing policies.
comment: 37 pages, 5 figures
☆ Optimization of Link Configuration for Satellite Communication Using Reinforcement Learning
Satellite communication is a key technology in our modern connected world. With increasingly complex hardware, one challenge is to efficiently configure links (connections) on a satellite transponder. Planning an optimal link configuration is extremely complex and depends on many parameters and metrics. The optimal use of the limited resources, bandwidth and power of the transponder is crucial. Such an optimization problem can be approximated using metaheuristic methods such as simulated annealing, but recent research results also show that reinforcement learning can achieve comparable or even better performance in optimization methods. However, there have not yet been any studies on link configuration on satellite transponders. In order to close this research gap, a transponder environment was developed as part of this work. For this environment, the performance of the reinforcement learning algorithm PPO was compared with the metaheuristic simulated annealing in two experiments. The results show that Simulated Annealing delivers better results for this static problem than the PPO algorithm, however, the research in turn also underlines the potential of reinforcement learning for optimization problems.
☆ ASTRID -- An Automated and Scalable TRIaD for the Evaluation of RAG-based Clinical Question Answering Systems
Large Language Models (LLMs) have shown impressive potential in clinical question answering (QA), with Retrieval Augmented Generation (RAG) emerging as a leading approach for ensuring the factual accuracy of model responses. However, current automated RAG metrics perform poorly in clinical and conversational use cases. Using clinical human evaluations of responses is expensive, unscalable, and not conducive to the continuous iterative development of RAG systems. To address these challenges, we introduce ASTRID - an Automated and Scalable TRIaD for evaluating clinical QA systems leveraging RAG - consisting of three metrics: Context Relevance (CR), Refusal Accuracy (RA), and Conversational Faithfulness (CF). Our novel evaluation metric, CF, is designed to better capture the faithfulness of a model's response to the knowledge base without penalising conversational elements. To validate our triad, we curate a dataset of over 200 real-world patient questions posed to an LLM-based QA agent during surgical follow-up for cataract surgery - the highest volume operation in the world - augmented with clinician-selected questions for emergency, clinical, and non-clinical out-of-domain scenarios. We demonstrate that CF can predict human ratings of faithfulness better than existing definitions for conversational use cases. Furthermore, we show that evaluation using our triad consisting of CF, RA, and CR exhibits alignment with clinician assessment for inappropriate, harmful, or unhelpful responses. Finally, using nine different LLMs, we demonstrate that the three metrics can closely agree with human evaluations, highlighting the potential of these metrics for use in LLM-driven automated evaluation pipelines. We also publish the prompts and datasets for these experiments, providing valuable resources for further research and development.
comment: 29 pages
☆ Modeling Feature Maps for Quantum Machine Learning
Quantum Machine Learning (QML) offers significant potential for complex tasks like genome sequence classification, but quantum noise on Noisy Intermediate-Scale Quantum (NISQ) devices poses practical challenges. This study systematically evaluates how various quantum noise models including dephasing, amplitude damping, depolarizing, thermal noise, bit-flip, and phase-flip affect key QML algorithms (QSVC, Peg-QSVC, QNN, VQC) and feature mapping techniques (ZFeatureMap, ZZFeatureMap, and PauliFeatureMap). Results indicate that QSVC is notably robust under noise, whereas Peg-QSVC and QNN are more sensitive, particularly to depolarizing and amplitude-damping noise. The PauliFeatureMap is especially vulnerable, highlighting difficulties in maintaining accurate classification under noisy conditions. These findings underscore the critical importance of feature map selection and noise mitigation strategies in optimizing QML for genomic classification, with promising implications for personalized medicine.
☆ EmoNeXt: an Adapted ConvNeXt for Facial Emotion Recognition SP
Facial expressions play a crucial role in human communication serving as a powerful and impactful means to express a wide range of emotions. With advancements in artificial intelligence and computer vision, deep neural networks have emerged as effective tools for facial emotion recognition. In this paper, we propose EmoNeXt, a novel deep learning framework for facial expression recognition based on an adapted ConvNeXt architecture network. We integrate a Spatial Transformer Network (STN) to focus on feature-rich regions of the face and Squeeze-and-Excitation blocks to capture channel-wise dependencies. Moreover, we introduce a self-attention regularization term, encouraging the model to generate compact feature vectors. We demonstrate the superiority of our model over existing state-of-the-art deep learning models on the FER2013 dataset regarding emotion classification accuracy.
comment: 6 pages, 5 figures and 2 tables. 2023 IEEE 25th International Workshop on Multimedia Signal Processing (MMSP), Poitiers, France
☆ PRESERVE: Prefetching Model Weights and KV-Cache in Distributed LLM Serving
Large language models (LLMs) are widely used across various applications, but their substantial computational requirements pose significant challenges, particularly in terms of HBM bandwidth bottlenecks and inter-device communication overhead. In this paper, we present PRESERVE, a novel prefetching framework designed to optimize LLM inference by overlapping memory reads for model weights and KV-cache with collective communication operations. Through extensive experiments conducted on commercial AI accelerators, we demonstrate up to 1.6x end-to-end speedup on state-of-the-art, open-source LLMs. Additionally, we perform a design space exploration that identifies the optimal hardware configuration for the proposed method, showing a further 1.25x improvement in performance per cost by selecting the optimal L2 cache size. Our results show that PRESERVE has the potential to mitigate the memory bottlenecks and communication overheads, offering a solution to improve the performance and scalability of the LLM inference systems.
☆ A Critical Synthesis of Uncertainty Quantification and Foundation Models in Monocular Depth Estimation
While recent foundation models have enabled significant breakthroughs in monocular depth estimation, a clear path towards safe and reliable deployment in the real-world remains elusive. Metric depth estimation, which involves predicting absolute distances, poses particular challenges, as even the most advanced foundation models remain prone to critical errors. Since quantifying the uncertainty has emerged as a promising endeavor to address these limitations and enable trustworthy deployment, we fuse five different uncertainty quantification methods with the current state-of-the-art DepthAnythingV2 foundation model. To cover a wide range of metric depth domains, we evaluate their performance on four diverse datasets. Our findings identify fine-tuning with the Gaussian Negative Log-Likelihood Loss (GNLL) as a particularly promising approach, offering reliable uncertainty estimates while maintaining predictive performance and computational efficiency on par with the baseline, encompassing both training and inference time. By fusing uncertainty quantification and foundation models within the context of monocular depth estimation, this paper lays a critical foundation for future research aimed at improving not only model performance but also its explainability. Extending this critical synthesis of uncertainty quantification and foundation models into other crucial tasks, such as semantic segmentation and pose estimation, presents exciting opportunities for safer and more reliable machine vision systems.
☆ A Multi-Modal AI Copilot for Single-Cell Analysis with Instruction Following
Large language models excel at interpreting complex natural language instructions, enabling them to perform a wide range of tasks. In the life sciences, single-cell RNA sequencing (scRNA-seq) data serves as the "language of cellular biology", capturing intricate gene expression patterns at the single-cell level. However, interacting with this "language" through conventional tools is often inefficient and unintuitive, posing challenges for researchers. To address these limitations, we present InstructCell, a multi-modal AI copilot that leverages natural language as a medium for more direct and flexible single-cell analysis. We construct a comprehensive multi-modal instruction dataset that pairs text-based instructions with scRNA-seq profiles from diverse tissues and species. Building on this, we develop a multi-modal cell language architecture capable of simultaneously interpreting and processing both modalities. InstructCell empowers researchers to accomplish critical tasks-such as cell type annotation, conditional pseudo-cell generation, and drug sensitivity prediction-using straightforward natural language commands. Extensive evaluations demonstrate that InstructCell consistently meets or exceeds the performance of existing single-cell foundation models, while adapting to diverse experimental conditions. More importantly, InstructCell provides an accessible and intuitive tool for exploring complex single-cell data, lowering technical barriers and enabling deeper biological insights.
comment: 37 pages; 13 figures; Code: https://github.com/zjunlp/Instructcell; Models: https://huggingface.co/zjunlp/Instructcell-chat, https://huggingface.co/zjunlp/InstructCell-instruct
☆ Assessing AI Adoption and Digitalization in SMEs: A Framework for Implementation
The primary objective of this research is to examine the current state of digitalization and the integration of artificial intelligence (AI) within small and medium-sized enterprises (SMEs) in Italy. There is a significant gap between SMEs and large corporations in their use of AI, with SMEs facing numerous barriers to adoption. This study identifies critical drivers and obstacles to achieving intelligent transformation, proposing a framework model to address key challenges and provide actionable guidelines
☆ CG-MER: A Card Game-based Multimodal dataset for Emotion Recognition
The field of affective computing has seen significant advancements in exploring the relationship between emotions and emerging technologies. This paper presents a novel and valuable contribution to this field with the introduction of a comprehensive French multimodal dataset designed specifically for emotion recognition. The dataset encompasses three primary modalities: facial expressions, speech, and gestures, providing a holistic perspective on emotions. Moreover, the dataset has the potential to incorporate additional modalities, such as Natural Language Processing (NLP) to expand the scope of emotion recognition research. The dataset was curated through engaging participants in card game sessions, where they were prompted to express a range of emotions while responding to diverse questions. The study included 10 sessions with 20 participants (9 females and 11 males). The dataset serves as a valuable resource for furthering research in emotion recognition and provides an avenue for exploring the intricate connections between human emotions and digital technologies.
comment: 8 pages, 2 figures and 4 tables. Sixteenth International Conference on Machine Vision (ICMV 2023), Yerevan, Armenia
☆ Revolutionizing Communication with Deep Learning and XAI for Enhanced Arabic Sign Language Recognition
This study introduces an integrated approach to recognizing Arabic Sign Language (ArSL) using state-of-the-art deep learning models such as MobileNetV3, ResNet50, and EfficientNet-B2. These models are further enhanced by explainable AI (XAI) techniques to boost interpretability. The ArSL2018 and RGB Arabic Alphabets Sign Language (AASL) datasets are employed, with EfficientNet-B2 achieving peak accuracies of 99.48\% and 98.99\%, respectively. Key innovations include sophisticated data augmentation methods to mitigate class imbalance, implementation of stratified 5-fold cross-validation for better generalization, and the use of Grad-CAM for clear model decision transparency. The proposed system not only sets new benchmarks in recognition accuracy but also emphasizes interpretability, making it suitable for applications in healthcare, education, and inclusive communication technologies.
comment: 13 pages, 25 figures, 16 tables
☆ LeapVAD: A Leap in Autonomous Driving via Cognitive Perception and Dual-Process Thinking
While autonomous driving technology has made remarkable strides, data-driven approaches still struggle with complex scenarios due to their limited reasoning capabilities. Meanwhile, knowledge-driven autonomous driving systems have evolved considerably with the popularization of visual language models. In this paper, we propose LeapVAD, a novel method based on cognitive perception and dual-process thinking. Our approach implements a human-attentional mechanism to identify and focus on critical traffic elements that influence driving decisions. By characterizing these objects through comprehensive attributes - including appearance, motion patterns, and associated risks - LeapVAD achieves more effective environmental representation and streamlines the decision-making process. Furthermore, LeapVAD incorporates an innovative dual-process decision-making module miming the human-driving learning process. The system consists of an Analytic Process (System-II) that accumulates driving experience through logical reasoning and a Heuristic Process (System-I) that refines this knowledge via fine-tuning and few-shot learning. LeapVAD also includes reflective mechanisms and a growing memory bank, enabling it to learn from past mistakes and continuously improve its performance in a closed-loop environment. To enhance efficiency, we develop a scene encoder network that generates compact scene representations for rapid retrieval of relevant driving experiences. Extensive evaluations conducted on two leading autonomous driving simulators, CARLA and DriveArena, demonstrate that LeapVAD achieves superior performance compared to camera-only approaches despite limited training data. Comprehensive ablation studies further emphasize its effectiveness in continuous learning and domain adaptation. Project page: https://pjlab-adg.github.io/LeapVAD/.
☆ Potential and Perils of Large Language Models as Judges of Unstructured Textual Data
Rapid advancements in large language models have unlocked remarkable capabilities when it comes to processing and summarizing unstructured text data. This has implications for the analysis of rich, open-ended datasets, such as survey responses, where LLMs hold the promise of efficiently distilling key themes and sentiments. However, as organizations increasingly turn to these powerful AI systems to make sense of textual feedback, a critical question arises, can we trust LLMs to accurately represent the perspectives contained within these text based datasets? While LLMs excel at generating human-like summaries, there is a risk that their outputs may inadvertently diverge from the true substance of the original responses. Discrepancies between the LLM-generated outputs and the actual themes present in the data could lead to flawed decision-making, with far-reaching consequences for organizations. This research investigates the effectiveness of LLMs as judge models to evaluate the thematic alignment of summaries generated by other LLMs. We utilized an Anthropic Claude model to generate thematic summaries from open-ended survey responses, with Amazon's Titan Express, Nova Pro, and Meta's Llama serving as LLM judges. The LLM-as-judge approach was compared to human evaluations using Cohen's kappa, Spearman's rho, and Krippendorff's alpha, validating a scalable alternative to traditional human centric evaluation methods. Our findings reveal that while LLMs as judges offer a scalable solution comparable to human raters, humans may still excel at detecting subtle, context-specific nuances. This research contributes to the growing body of knowledge on AI assisted text analysis. We discuss limitations and provide recommendations for future research, emphasizing the need for careful consideration when generalizing LLM judge models across various contexts and use cases.
comment: 11 pages, 1 appendix
☆ I Can Find You in Seconds! Leveraging Large Language Models for Code Authorship Attribution
Source code authorship attribution is important in software forensics, plagiarism detection, and protecting software patch integrity. Existing techniques often rely on supervised machine learning, which struggles with generalization across different programming languages and coding styles due to the need for large labeled datasets. Inspired by recent advances in natural language authorship analysis using large language models (LLMs), which have shown exceptional performance without task-specific tuning, this paper explores the use of LLMs for source code authorship attribution. We present a comprehensive study demonstrating that state-of-the-art LLMs can successfully attribute source code authorship across different languages. LLMs can determine whether two code snippets are written by the same author with zero-shot prompting, achieving a Matthews Correlation Coefficient (MCC) of 0.78, and can attribute code authorship from a small set of reference code snippets via few-shot learning, achieving MCC of 0.77. Additionally, LLMs show some adversarial robustness against misattribution attacks. Despite these capabilities, we found that naive prompting of LLMs does not scale well with a large number of authors due to input token limitations. To address this, we propose a tournament-style approach for large-scale attribution. Evaluating this approach on datasets of C++ (500 authors, 26,355 samples) and Java (686 authors, 55,267 samples) code from GitHub, we achieve classification accuracy of up to 65% for C++ and 68.7% for Java using only one reference per author. These results open new possibilities for applying LLMs to code authorship attribution in cybersecurity and software engineering.
comment: 12 pages, 5 figures,
☆ FairTTTS: A Tree Test Time Simulation Method for Fairness-Aware Classification
Algorithmic decision-making has become deeply ingrained in many domains, yet biases in machine learning models can still produce discriminatory outcomes, often harming unprivileged groups. Achieving fair classification is inherently challenging, requiring a careful balance between predictive performance and ethical considerations. We present FairTTTS, a novel post-processing bias mitigation method inspired by the Tree Test Time Simulation (TTTS) method. Originally developed to enhance accuracy and robustness against adversarial inputs through probabilistic decision-path adjustments, TTTS serves as the foundation for FairTTTS. By building on this accuracy-enhancing technique, FairTTTS mitigates bias and improves predictive performance. FairTTTS uses a distance-based heuristic to adjust decisions at protected attribute nodes, ensuring fairness for unprivileged samples. This fairness-oriented adjustment occurs as a post-processing step, allowing FairTTTS to be applied to pre-trained models, diverse datasets, and various fairness metrics without retraining. Extensive evaluation on seven benchmark datasets shows that FairTTTS outperforms traditional methods in fairness improvement, achieving a 20.96% average increase over the baseline compared to 18.78% for related work, and further enhances accuracy by 0.55%. In contrast, competing methods typically reduce accuracy by 0.42%. These results confirm that FairTTTS effectively promotes more equitable decision-making while simultaneously improving predictive performance.
☆ Multiple-Input Variational Auto-Encoder for Anomaly Detection in Heterogeneous Data
Anomaly detection (AD) plays a pivotal role in AI applications, e.g., in classification, and intrusion/threat detection in cybersecurity. However, most existing methods face challenges of heterogeneity amongst feature subsets posed by non-independent and identically distributed (non-IID) data. We propose a novel neural network model called Multiple-Input Auto-Encoder for AD (MIAEAD) to address this. MIAEAD assigns an anomaly score to each feature subset of a data sample to indicate its likelihood of being an anomaly. This is done by using the reconstruction error of its sub-encoder as the anomaly score. All sub-encoders are then simultaneously trained using unsupervised learning to determine the anomaly scores of feature subsets. The final AUC of MIAEAD is calculated for each sub-dataset, and the maximum AUC obtained among the sub-datasets is selected. To leverage the modelling of the distribution of normal data to identify anomalies of the generative models, we develop a novel neural network architecture/model called Multiple-Input Variational Auto-Encoder (MIVAE). MIVAE can process feature subsets through its sub-encoders before learning distribution of normal data in the latent space. This allows MIVAE to identify anomalies that deviate from the learned distribution. We theoretically prove that the difference in the average anomaly score between normal samples and anomalies obtained by the proposed MIVAE is greater than that of the Variational Auto-Encoder (VAEAD), resulting in a higher AUC for MIVAE. Extensive experiments on eight real-world anomaly datasets demonstrate the superior performance of MIAEAD and MIVAE over conventional methods and the state-of-the-art unsupervised models, by up to 6% in terms of AUC score. Alternatively, MIAEAD and MIVAE have a high AUC when applied to feature subsets with low heterogeneity based on the coefficient of variation (CV) score.
comment: 16 pages
☆ Refusal Behavior in Large Language Models: A Nonlinear Perspective
Refusal behavior in large language models (LLMs) enables them to decline responding to harmful, unethical, or inappropriate prompts, ensuring alignment with ethical standards. This paper investigates refusal behavior across six LLMs from three architectural families. We challenge the assumption of refusal as a linear phenomenon by employing dimensionality reduction techniques, including PCA, t-SNE, and UMAP. Our results reveal that refusal mechanisms exhibit nonlinear, multidimensional characteristics that vary by model architecture and layer. These findings highlight the need for nonlinear interpretability to improve alignment research and inform safer AI deployment strategies.
☆ EEG-ReMinD: Enhancing Neurodegenerative EEG Decoding through Self-Supervised State Reconstruction-Primed Riemannian Dynamics
The development of EEG decoding algorithms confronts challenges such as data sparsity, subject variability, and the need for precise annotations, all of which are vital for advancing brain-computer interfaces and enhancing the diagnosis of diseases. To address these issues, we propose a novel two-stage approach named Self-Supervised State Reconstruction-Primed Riemannian Dynamics (EEG-ReMinD) , which mitigates reliance on supervised learning and integrates inherent geometric features. This approach efficiently handles EEG data corruptions and reduces the dependency on labels. EEG-ReMinD utilizes self-supervised and geometric learning techniques, along with an attention mechanism, to analyze the temporal dynamics of EEG features within the framework of Riemannian geometry, referred to as Riemannian dynamics. Comparative analyses on both intact and corrupted datasets from two different neurodegenerative disorders underscore the enhanced performance of EEG-ReMinD.
☆ An Empirical Wall-Pressure Spectrum Model for Aeroacoustic Predictions Based on Symbolic Regression
Fast-turn around methods to predict airfoil trailing-edge noise are crucial for incorporating noise limitations into design optimization loops of several applications. Among these aeroacoustic predictive models, Amiet's theory offers the best balance between accuracy and simplicity. The accuracy of the model relies heavily on precise wall-pressure spectrum predictions, which are often based on single-equation formulations with adjustable parameters. These parameters are calibrated for particular airfoils and flow conditions and consequently tend to fail when applied outside their calibration range. This paper introduces a new wall-pressure spectrum empirical model designed to enhance the robustness and accuracy of current state-of-the-art predictions while widening the range of applicability of the model to different airfoils and flow conditions. The model is developed using AI-based symbolic regression via a genetic-algorithm-based approach, and applied to a dataset of wall-pressure fluctuations measured on NACA 0008 and NACA 63018 airfoils at multiple angles of attack and inflow velocities, covering turbulent boundary layers with both adverse and favorable pressure gradients. Validation against experimental data (outside the training dataset) demonstrates the robustness of the model compared to well-accepted semi-empirical models. Finally, the model is integrated with Amiet's theory to predict the aeroacoustic noise of a full-scale wind turbine, showing good agreement with experimental measurements.
☆ In-situ graph reasoning and knowledge expansion using Graph-PReFLexOR
The pursuit of automated scientific discovery has fueled progress from symbolic logic to modern AI, forging new frontiers in reasoning and pattern recognition. Transformers function as potential systems, where every possible relationship remains latent potentiality until tasks impose constraints, akin to measurement. Yet, refining their sampling requires more than probabilistic selection: solutions must conform to specific structures or rules, ensuring consistency and the invocation of general principles. We present Graph-PReFLexOR (Graph-based Preference-based Recursive Language Modeling for Exploratory Optimization of Reasoning), a framework that combines graph reasoning with symbolic abstraction to dynamically expand domain knowledge. Inspired by reinforcement learning, Graph-PReFLexOR defines reasoning as a structured mapping, where tasks yield knowledge graphs, abstract patterns, and ultimately, final answers. Inspired by category theory, it encodes concepts as nodes and their relationships as edges, supporting hierarchical inference and adaptive learning through isomorphic representations. Demonstrations include hypothesis generation, materials design, and creative reasoning, such as discovering relationships between mythological concepts like 'thin places' with materials science. We propose a 'knowledge garden growth' strategy that integrates insights across domains, promoting interdisciplinary connections. Results with a 3-billion-parameter Graph-PReFLexOR model show superior reasoning depth and adaptability, underscoring the potential for transparent, multidisciplinary AI-driven discovery. It lays the groundwork for general autonomous reasoning solutions.
☆ Data-driven inventory management for new products: A warm-start and adjusted Dyna-$Q$ approach
In this paper, we propose a novel reinforcement learning algorithm for inventory management of newly launched products with no or limited historical demand information. The algorithm follows the classic Dyna-$Q$ structure, balancing the model-based and model-free approaches, while accelerating the training process of Dyna-$Q$ and mitigating the model discrepancy generated by the model-based feedback. Warm-start information from the demand data of existing similar products can be incorporated into the algorithm to further stabilize the early-stage training and reduce the variance of the estimated optimal policy. Our approach is validated through a case study of bakery inventory management with real data. The adjusted Dyna-$Q$ shows up to a 23.7\% reduction in average daily cost compared with $Q$-learning, and up to a 77.5\% reduction in training time within the same horizon compared with classic Dyna-$Q$. By incorporating the warm-start information, it can be found that the adjusted Dyna-$Q$ has the lowest total cost, lowest variance in total cost, and relatively low shortage percentages among all the algorithms under a 30-day testing.
comment: 7 pages, 2 figures
☆ Consistency of Responses and Continuations Generated by Large Language Models on Social Media
Large Language Models (LLMs) demonstrate remarkable capabilities in text generation, yet their emotional consistency and semantic coherence in social media contexts remain insufficiently understood. This study investigates how LLMs handle emotional content and maintain semantic relationships through continuation and response tasks using two open-source models: Gemma and Llama. By analyzing climate change discussions from Twitter and Reddit, we examine emotional transitions, intensity patterns, and semantic similarity between human-authored and LLM-generated content. Our findings reveal that while both models maintain high semantic coherence, they exhibit distinct emotional patterns: Gemma shows a tendency toward negative emotion amplification, particularly anger, while maintaining certain positive emotions like optimism. Llama demonstrates superior emotional preservation across a broader spectrum of affects. Both models systematically generate responses with attenuated emotional intensity compared to human-authored content and show a bias toward positive emotions in response tasks. Additionally, both models maintain strong semantic similarity with original texts, though performance varies between continuation and response tasks. These findings provide insights into LLMs' emotional and semantic processing capabilities, with implications for their deployment in social media contexts and human-AI interaction design.
☆ Guiding the classification of hepatocellular carcinoma on 3D CT-scans using deep and handcrafted radiological features
Hepatocellular carcinoma is the most spread primary liver cancer across the world ($\sim$80\% of the liver tumors). The gold standard for HCC diagnosis is liver biopsy. However, in the clinical routine, expert radiologists provide a visual diagnosis by interpreting hepatic CT-scans according to a standardized protocol, the LI-RADS, which uses five radiological criteria with an associated decision tree. In this paper, we propose an automatic approach to predict histology-proven HCC from CT images in order to reduce radiologists' inter-variability. We first show that standard deep learning methods fail to accurately predict HCC from CT-scans on a challenging database, and propose a two-step approach inspired by the LI-RADS system to improve the performance. We achieve improvements from 6 to 18 points of AUC with respect to deep learning baselines trained with different architectures. We also provide clinical validation of our method, achieving results that outperform non-expert radiologists and are on par with expert ones.
comment: IEEE ISBI 2025
☆ Hybrid Action Based Reinforcement Learning for Multi-Objective Compatible Autonomous Driving
Reinforcement Learning (RL) has shown excellent performance in solving decision-making and control problems of autonomous driving, which is increasingly applied in diverse driving scenarios. However, driving is a multi-attribute problem, leading to challenges in achieving multi-objective compatibility for current RL methods, especially in both policy execution and policy iteration. On the one hand, the common action space structure with single action type limits driving flexibility or results in large behavior fluctuations during policy execution. On the other hand, the multi-attribute weighted single reward function result in the agent's disproportionate attention to certain objectives during policy iterations. To this end, we propose a Multi-objective Ensemble-Critic reinforcement learning method with Hybrid Parametrized Action for multi-objective compatible autonomous driving. Specifically, a parameterized action space is constructed to generate hybrid driving actions, combining both abstract guidance and concrete control commands. A multi-objective critics architecture is constructed considering multiple attribute rewards, to ensure simultaneously focusing on different driving objectives. Additionally, uncertainty-based exploration strategy is introduced to help the agent faster approach viable driving policy. The experimental results in both the simulated traffic environment and the HighD dataset demonstrate that our method can achieve multi-objective compatible autonomous driving in terms of driving efficiency, action consistency, and safety. It enhances the general performance of the driving while significantly increasing training efficiency.
comment: 12 pages, 9 figures, 5 tables
☆ Hierarchical Autoscaling for Large Language Model Serving with Chiron
Large language model (LLM) serving is becoming an increasingly important workload for cloud providers. Based on performance SLO requirements, LLM inference requests can be divided into (a) interactive requests that have tight SLOs in the order of seconds, and (b) batch requests that have relaxed SLO in the order of minutes to hours. These SLOs can degrade based on the arrival rates, multiplexing, and configuration parameters, thus necessitating the use of resource autoscaling on serving instances and their batch sizes. However, previous autoscalers for LLM serving do not consider request SLOs leading to unnecessary scaling and resource under-utilization. To address these limitations, we introduce Chiron, an autoscaler that uses the idea of hierarchical backpressure estimated using queue size, utilization, and SLOs. Our experiments show that Chiron achieves up to 90% higher SLO attainment and improves GPU efficiency by up to 70% compared to existing solutions.
☆ NOMTO: Neural Operator-based symbolic Model approximaTion and discOvery
While many physical and engineering processes are most effectively described by non-linear symbolic models, existing non-linear symbolic regression (SR) methods are restricted to a limited set of continuous algebraic functions, thereby limiting their applicability to discover higher order non-linear differential relations. In this work, we introduce the Neural Operator-based symbolic Model approximaTion and discOvery (NOMTO) method, a novel approach to symbolic model discovery that leverages Neural Operators to encompass a broad range of symbolic operations. We demonstrate that NOMTO can successfully identify symbolic expressions containing elementary functions with singularities, special functions, and derivatives. Additionally, our experiments demonstrate that NOMTO can accurately rediscover second-order non-linear partial differential equations. By broadening the set of symbolic operations available for discovery, NOMTO significantly advances the capabilities of existing SR methods. It provides a powerful and flexible tool for model discovery, capable of capturing complex relations in a variety of physical systems.
☆ Artificial Liver Classifier: A New Alternative to Conventional Machine Learning Models
Supervised machine learning classifiers often encounter challenges related to performance, accuracy, and overfitting. This paper introduces the Artificial Liver Classifier (ALC), a novel supervised learning classifier inspired by the human liver's detoxification function. The ALC is characterized by its simplicity, speed, hyperparameters-free, ability to reduce overfitting, and effectiveness in addressing multi-classification problems through straightforward mathematical operations. To optimize the ALC's parameters, an improved FOX optimization algorithm (IFOX) is employed as the training method. The proposed ALC was evaluated on five benchmark machine learning datasets: Iris Flower, Breast Cancer Wisconsin, Wine, Voice Gender, and MNIST. The results demonstrated competitive performance, with the ALC achieving 100% accuracy on the Iris dataset, surpassing logistic regression, multilayer perceptron, and support vector machine. Similarly, on the Breast Cancer dataset, it achieved 99.12% accuracy, outperforming XGBoost and logistic regression. Across all datasets, the ALC consistently exhibited lower overfitting gaps and loss compared to conventional classifiers. These findings highlight the potential of leveraging biological process simulations to develop efficient machine learning models and open new avenues for innovation in the field.
comment: 21 pages
☆ A Roadmap to Guide the Integration of LLMs in Hierarchical Planning AAAI
Recent advances in Large Language Models (LLMs) are fostering their integration into several reasoning-related fields, including Automated Planning (AP). However, their integration into Hierarchical Planning (HP), a subfield of AP that leverages hierarchical knowledge to enhance planning performance, remains largely unexplored. In this preliminary work, we propose a roadmap to address this gap and harness the potential of LLMs for HP. To this end, we present a taxonomy of integration methods, exploring how LLMs can be utilized within the HP life cycle. Additionally, we provide a benchmark with a standardized dataset for evaluating the performance of future LLM-based HP approaches, and present initial results for a state-of-the-art HP planner and LLM planner. As expected, the latter exhibits limited performance (3\% correct plans, and none with a correct hierarchical decomposition) but serves as a valuable baseline for future approaches.
comment: 5 pages, 0 figures, to be published in the AAAI Workshop on Planning in the Era of LLMs ( https://llmforplanning.github.io )
☆ Optimizing Speech Multi-View Feature Fusion through Conditional Computation ICASSP 2025
Recent advancements have highlighted the efficacy of self-supervised learning (SSL) features in various speech-related tasks, providing lightweight and versatile multi-view speech representations. However, our study reveals that while SSL features expedite model convergence, they conflict with traditional spectral features like FBanks in terms of update directions. In response, we propose a novel generalized feature fusion framework grounded in conditional computation, featuring a gradient-sensitive gating network and a multi-stage dropout strategy. This framework mitigates feature conflicts and bolsters model robustness to multi-view input features. By integrating SSL and spectral features, our approach accelerates convergence and maintains performance on par with spectral models across multiple speech translation tasks on the MUSTC dataset.
comment: ICASSP 2025
☆ Exploring Narrative Clustering in Large Language Models: A Layerwise Analysis of BERT
This study investigates the internal mechanisms of BERT, a transformer-based large language model, with a focus on its ability to cluster narrative content and authorial style across its layers. Using a dataset of narratives developed via GPT-4, featuring diverse semantic content and stylistic variations, we analyze BERT's layerwise activations to uncover patterns of localized neural processing. Through dimensionality reduction techniques such as Principal Component Analysis (PCA) and Multidimensional Scaling (MDS), we reveal that BERT exhibits strong clustering based on narrative content in its later layers, with progressively compact and distinct clusters. While strong stylistic clustering might occur when narratives are rephrased into different text types (e.g., fables, sci-fi, kids' stories), minimal clustering is observed for authorial style specific to individual writers. These findings highlight BERT's prioritization of semantic content over stylistic features, offering insights into its representational capabilities and processing hierarchy. This study contributes to understanding how transformer models like BERT encode linguistic information, paving the way for future interdisciplinary research in artificial intelligence and cognitive neuroscience.
comment: arXiv admin note: text overlap with arXiv:2408.03062, arXiv:2408.04270, arXiv:2307.01577
☆ Self-Attentive Spatio-Temporal Calibration for Precise Intermediate Layer Matching in ANN-to-SNN Distillation
Spiking Neural Networks (SNNs) are promising for low-power computation due to their event-driven mechanism but often suffer from lower accuracy compared to Artificial Neural Networks (ANNs). ANN-to-SNN knowledge distillation can improve SNN performance, but previous methods either focus solely on label information, missing valuable intermediate layer features, or use a layer-wise approach that neglects spatial and temporal semantic inconsistencies, leading to performance degradation.To address these limitations, we propose a novel method called self-attentive spatio-temporal calibration (SASTC). SASTC uses self-attention to identify semantically aligned layer pairs between ANN and SNN, both spatially and temporally. This enables the autonomous transfer of relevant semantic information. Extensive experiments show that SASTC outperforms existing methods, effectively solving the mismatching problem. Superior accuracy results include 95.12% on CIFAR-10, 79.40% on CIFAR-100 with 2 time steps, and 68.69% on ImageNet with 4 time steps for static datasets, and 97.92% on DVS-Gesture and 83.60% on DVS-CIFAR10 for neuromorphic datasets. This marks the first time SNNs have outperformed ANNs on both CIFAR-10 and CIFAR-100, shedding the new light on the potential applications of SNNs.
☆ Building Symbiotic AI: Reviewing the AI Act for a Human-Centred, Principle-Based Framework
Artificial Intelligence (AI) spreads quickly as new technologies and services take over modern society. The need to regulate AI design, development, and use is strictly necessary to avoid unethical and potentially dangerous consequences to humans. The European Union (EU) has released a new legal framework, the AI Act, to regulate AI by undertaking a risk-based approach to safeguard humans during interaction. At the same time, researchers offer a new perspective on AI systems, commonly known as Human-Centred AI (HCAI), highlighting the need for a human-centred approach to their design. In this context, Symbiotic AI (a subtype of HCAI) promises to enhance human capabilities through a deeper and continuous collaboration between human intelligence and AI. This article presents the results of a Systematic Literature Review (SLR) that aims to identify principles that characterise the design and development of Symbiotic AI systems while considering humans as the core of the process. Through content analysis, four principles emerged from the review that must be applied to create Human-Centred AI systems that can establish a symbiotic relationship with humans. In addition, current trends and challenges were defined to indicate open questions that may guide future research for the development of SAI systems that comply with the AI Act.
comment: First version: 17 pages, 5 figures, 2 tables
☆ Exploring visual language models as a powerful tool in the diagnosis of Ewing Sarcoma
Ewing's sarcoma (ES), characterized by a high density of small round blue cells without structural organization, presents a significant health concern, particularly among adolescents aged 10 to 19. Artificial intelligence-based systems for automated analysis of histopathological images are promising to contribute to an accurate diagnosis of ES. In this context, this study explores the feature extraction ability of different pre-training strategies for distinguishing ES from other soft tissue or bone sarcomas with similar morphology in digitized tissue microarrays for the first time, as far as we know. Vision-language supervision (VLS) is compared to fully-supervised ImageNet pre-training within a multiple instance learning paradigm. Our findings indicate a substantial improvement in diagnostic accuracy with the adaption of VLS using an in-domain dataset. Notably, these models not only enhance the accuracy of predicted classes but also drastically reduce the number of trainable parameters and computational costs.
comment: 11 pages, 5 figures, 2 tables. Oral presentation at KES-InMed 2024 held in Madeira, Portugal
☆ READ: Reinforcement-based Adversarial Learning for Text Classification with Limited Labeled Data
Pre-trained transformer models such as BERT have shown massive gains across many text classification tasks. However, these models usually need enormous labeled data to achieve impressive performances. Obtaining labeled data is often expensive and time-consuming, whereas collecting unlabeled data using some heuristics is relatively much cheaper for any task. Therefore, this paper proposes a method that encapsulates reinforcement learning-based text generation and semi-supervised adversarial learning approaches in a novel way to improve the model's performance. Our method READ, Reinforcement-based Adversarial learning, utilizes an unlabeled dataset to generate diverse synthetic text through reinforcement learning, improving the model's generalization capability using adversarial learning. Our experimental results show that READ outperforms the existing state-of-art methods on multiple datasets.
☆ Cooperative Patrol Routing: Optimizing Urban Crime Surveillance through Multi-Agent Reinforcement Learning
The effective design of patrol strategies is a difficult and complex problem, especially in medium and large areas. The objective is to plan, in a coordinated manner, the optimal routes for a set of patrols in a given area, in order to achieve maximum coverage of the area, while also trying to minimize the number of patrols. In this paper, we propose a multi-agent reinforcement learning (MARL) model, based on a decentralized partially observable Markov decision process, to plan unpredictable patrol routes within an urban environment represented as an undirected graph. The model attempts to maximize a target function that characterizes the environment within a given time frame. Our model has been tested to optimize police patrol routes in three medium-sized districts of the city of Malaga. The aim was to maximize surveillance coverage of the most crime-prone areas, based on actual crime data in the city. To address this problem, several MARL algorithms have been studied, and among these the Value Decomposition Proximal Policy Optimization (VDPPO) algorithm exhibited the best performance. We also introduce a novel metric, the coverage index, for the evaluation of the coverage performance of the routes generated by our model. This metric is inspired by the predictive accuracy index (PAI), which is commonly used in criminology to detect hotspots. Using this metric, we have evaluated the model under various scenarios in which the number of agents (or patrols), their starting positions, and the level of information they can observe in the environment have been modified. Results show that the coordinated routes generated by our model achieve a coverage of more than $90\%$ of the $3\%$ of graph nodes with the highest crime incidence, and $65\%$ for $20\%$ of these nodes; $3\%$ and $20\%$ represent the coverage standards for police resource allocation.
☆ An AI-driven framework for rapid and localized optimizations of urban open spaces
As urbanization accelerates, open spaces are increasingly recognized for their role in enhancing sustainability and well-being, yet they remain underexplored compared to built spaces. This study introduces an AI-driven framework that integrates machine learning models (MLMs) and explainable AI techniques to optimize Sky View Factor (SVF) and visibility, key spatial metrics influencing thermal comfort and perceived safety in urban spaces. Unlike global optimization methods, which are computationally intensive and impractical for localized adjustments, this framework supports incremental design improvements with lower computational costs and greater flexibility. The framework employs SHapley Adaptive Explanations (SHAP) to analyze feature importance and Counterfactual Explanations (CFXs) to propose minimal design changes. Simulations tested five MLMs, identifying XGBoost as the most accurate, with building width, park area, and heights of surrounding buildings as critical for SVF, and distances from southern buildings as key for visibility. Compared to Genetic Algorithms, which required approximately 15/30 minutes across 3/4 generations to converge, the tested CFX approach achieved optimized results in 1 minute with a 5% RMSE error, demonstrating significantly faster performance and suitability for scalable retrofitting strategies. This interpretable and computationally efficient framework advances urban performance optimization, providing data-driven insights and practical retrofitting solutions for enhancing usability and environmental quality across diverse urban contexts.
comment: 36 pages
☆ Tutorial: VAE as an inference paradigm for neuroimaging
In this tutorial, we explore Variational Autoencoders (VAEs), an essential framework for unsupervised learning, particularly suited for high-dimensional datasets such as neuroimaging. By integrating deep learning with Bayesian inference, VAEs enable the generation of interpretable latent representations. This tutorial outlines the theoretical foundations of VAEs, addresses practical challenges such as convergence issues and over-fitting, and discusses strategies like the reparameterization trick and hyperparameter optimization. We also highlight key applications of VAEs in neuroimaging, demonstrating their potential to uncover meaningful patterns, including those associated with neurodegenerative processes, and their broader implications for analyzing complex brain data.
comment: 18 pages, 4 figures
☆ TriAdaptLoRA: Brain-Inspired Triangular Adaptive Low-Rank Adaptation for Parameter-Efficient Fine-Tuning
The fine-tuning of Large Language Models (LLMs) is pivotal for achieving optimal performance across diverse downstream tasks. However, while full fine-tuning delivers superior results, it entails significant computational and resource costs. Parameter-Efficient Fine-Tuning (PEFT) methods, such as LoRA, address these challenges by reducing the number of trainable parameters, but they often struggle with rank adjustment efficiency and task-specific adaptability. We propose Triangular Adaptive Low-Rank Adaptation (TriAdaptLoRA), a novel PEFT framework inspired by neuroscience principles, which dynamically optimizes the allocation of trainable parameters. TriAdaptLoRA introduces three key innovations: 1) a triangular split of transformation matrices into lower and upper triangular components to maximize parameter utilization, 2) a parameter importance metric based on normalized Frobenius norms for efficient adaptation, and 3) an adaptive rank-growth strategy governed by dynamic thresholds, allowing flexible parameter allocation across training steps. Experiments conducted on a variety of natural language understanding and generation tasks demonstrate that TriAdaptLoRA consistently outperforms existing PEFT methods. It achieves superior performance, enhanced stability, and reduced computational overhead, particularly under linear threshold-driven rank growth. These results highlight its efficacy as a scalable and resource-efficient solution for fine-tuning LLMs.
☆ DisCoPatch: Batch Statistics Are All You Need For OOD Detection, But Only If You Can Trust Them
Out-of-distribution (OOD) detection holds significant importance across many applications. While semantic and domain-shift OOD problems are well-studied, this work focuses on covariate shifts - subtle variations in the data distribution that can degrade machine learning performance. We hypothesize that detecting these subtle shifts can improve our understanding of in-distribution boundaries, ultimately improving OOD detection. In adversarial discriminators trained with Batch Normalization (BN), real and adversarial samples form distinct domains with unique batch statistics - a property we exploit for OOD detection. We introduce DisCoPatch, an unsupervised Adversarial Variational Autoencoder (VAE) framework that harnesses this mechanism. During inference, batches consist of patches from the same image, ensuring a consistent data distribution that allows the model to rely on batch statistics. DisCoPatch uses the VAE's suboptimal outputs (generated and reconstructed) as negative samples to train the discriminator, thereby improving its ability to delineate the boundary between in-distribution samples and covariate shifts. By tightening this boundary, DisCoPatch achieves state-of-the-art results in public OOD detection benchmarks. The proposed model not only excels in detecting covariate shifts, achieving 95.5% AUROC on ImageNet-1K(-C) but also outperforms all prior methods on public Near-OOD (95.0%) benchmarks. With a compact model size of 25MB, it achieves high OOD detection performance at notably lower latency than existing methods, making it an efficient and practical solution for real-world OOD detection applications. The code will be made publicly available
☆ Maximizing Uncertainty for Federated learning via Bayesian Optimisation-based Model Poisoning
As we transition from Narrow Artificial Intelligence towards Artificial Super Intelligence, users are increasingly concerned about their privacy and the trustworthiness of machine learning (ML) technology. A common denominator for the metrics of trustworthiness is the quantification of uncertainty inherent in DL algorithms, and specifically in the model parameters, input data, and model predictions. One of the common approaches to address privacy-related issues in DL is to adopt distributed learning such as federated learning (FL), where private raw data is not shared among users. Despite the privacy-preserving mechanisms in FL, it still faces challenges in trustworthiness. Specifically, the malicious users, during training, can systematically create malicious model parameters to compromise the models predictive and generative capabilities, resulting in high uncertainty about their reliability. To demonstrate malicious behaviour, we propose a novel model poisoning attack method named Delphi which aims to maximise the uncertainty of the global model output. We achieve this by taking advantage of the relationship between the uncertainty and the model parameters of the first hidden layer of the local model. Delphi employs two types of optimisation , Bayesian Optimisation and Least Squares Trust Region, to search for the optimal poisoned model parameters, named as Delphi-BO and Delphi-LSTR. We quantify the uncertainty using the KL Divergence to minimise the distance of the predictive probability distribution towards an uncertain distribution of model output. Furthermore, we establish a mathematical proof for the attack effectiveness demonstrated in FL. Numerical results demonstrate that Delphi-BO induces a higher amount of uncertainty than Delphi-LSTR highlighting vulnerability of FL systems to model poisoning attacks.
comment: 14 pages
☆ GDiffRetro: Retrosynthesis Prediction with Dual Graph Enhanced Molecular Representation and Diffusion Generation
Retrosynthesis prediction focuses on identifying reactants capable of synthesizing a target product. Typically, the retrosynthesis prediction involves two phases: Reaction Center Identification and Reactant Generation. However, we argue that most existing methods suffer from two limitations in the two phases: (i) Existing models do not adequately capture the ``face'' information in molecular graphs for the reaction center identification. (ii) Current approaches for the reactant generation predominantly use sequence generation in a 2D space, which lacks versatility in generating reasonable distributions for completed reactive groups and overlooks molecules' inherent 3D properties. To overcome the above limitations, we propose GDiffRetro. For the reaction center identification, GDiffRetro uniquely integrates the original graph with its corresponding dual graph to represent molecular structures, which helps guide the model to focus more on the faces in the graph. For the reactant generation, GDiffRetro employs a conditional diffusion model in 3D to further transform the obtained synthon into a complete reactant. Our experimental findings reveal that GDiffRetro outperforms state-of-the-art semi-template models across various evaluative metrics.
☆ LLM-Ehnanced Holonic Architecture for Ad-Hoc Scalable SoS
As modern system of systems (SoS) become increasingly adaptive and human centred, traditional architectures often struggle to support interoperability, reconfigurability, and effective human system interaction. This paper addresses these challenges by advancing the state of the art holonic architecture for SoS, offering two main contributions to support these adaptive needs. First, we propose a layered architecture for holons, which includes reasoning, communication, and capabilities layers. This design facilitates seamless interoperability among heterogeneous constituent systems by improving data exchange and integration. Second, inspired by principles of intelligent manufacturing, we introduce specialised holons namely, supervisor, planner, task, and resource holons aimed at enhancing the adaptability and reconfigurability of SoS. These specialised holons utilise large language models within their reasoning layers to support decision making and ensure real time adaptability. We demonstrate our approach through a 3D mobility case study focused on smart city transportation, showcasing its potential for managing complex, multimodal SoS environments. Additionally, we propose evaluation methods to assess the architecture efficiency and scalability,laying the groundwork for future empirical validations through simulations and real world implementations.
☆ Training Hybrid Neural Networks with Multimode Optical Nonlinearities Using Digital Twins
The ability to train ever-larger neural networks brings artificial intelligence to the forefront of scientific and technical discoveries. However, their exponentially increasing size creates a proportionally greater demand for energy and computational hardware. Incorporating complex physical events in networks as fixed, efficient computation modules can address this demand by decreasing the complexity of trainable layers. Here, we utilize ultrashort pulse propagation in multimode fibers, which perform large-scale nonlinear transformations, for this purpose. Training the hybrid architecture is achieved through a neural model that differentiably approximates the optical system. The training algorithm updates the neural simulator and backpropagates the error signal over this proxy to optimize layers preceding the optical one. Our experimental results achieve state-of-the-art image classification accuracies and simulation fidelity. Moreover, the framework demonstrates exceptional resilience to experimental drifts. By integrating low-energy physical systems into neural networks, this approach enables scalable, energy-efficient AI models with significantly reduced computational demands.
comment: 17 pages, 6 figures
☆ GAC-Net_Geometric and attention-based Network for Depth Completion
Depth completion is a key task in autonomous driving, aiming to complete sparse LiDAR depth measurements into high-quality dense depth maps through image guidance. However, existing methods usually treat depth maps as an additional channel of color images, or directly perform convolution on sparse data, failing to fully exploit the 3D geometric information in depth maps, especially with limited performance in complex boundaries and sparse areas. To address these issues, this paper proposes a depth completion network combining channel attention mechanism and 3D global feature perception (CGA-Net). The main innovations include: 1) Utilizing PointNet++ to extract global 3D geometric features from sparse depth maps, enhancing the scene perception ability of low-line LiDAR data; 2) Designing a channel-attention-based multimodal feature fusion module to efficiently integrate sparse depth, RGB images, and 3D geometric features; 3) Combining residual learning with CSPN++ to optimize the depth refinement stage, further improving the completion quality in edge areas and complex scenes. Experiments on the KITTI depth completion dataset show that CGA-Net can significantly improve the prediction accuracy of dense depth maps, achieving a new state-of-the-art (SOTA), and demonstrating strong robustness to sparse and complex scenes.
comment: 13pages,4 figures, 2 tables
☆ Facial Dynamics in Video: Instruction Tuning for Improved Facial Expression Perception and Contextual Awareness
Facial expression captioning has found widespread application across various domains. Recently, the emergence of video Multimodal Large Language Models (MLLMs) has shown promise in general video understanding tasks. However, describing facial expressions within videos poses two major challenges for these models: (1) the lack of adequate datasets and benchmarks, and (2) the limited visual token capacity of video MLLMs. To address these issues, this paper introduces a new instruction-following dataset tailored for dynamic facial expression caption. The dataset comprises 5,033 high-quality video clips annotated manually, containing over 700,000 tokens. Its purpose is to improve the capability of video MLLMs to discern subtle facial nuances. Furthermore, we propose FaceTrack-MM, which leverages a limited number of tokens to encode the main character's face. This model demonstrates superior performance in tracking faces and focusing on the facial expressions of the main characters, even in intricate multi-person scenarios. Additionally, we introduce a novel evaluation metric combining event extraction, relation classification, and the longest common subsequence (LCS) algorithm to assess the content consistency and temporal sequence consistency of generated text. Moreover, we present FEC-Bench, a benchmark designed to assess the performance of existing video MLLMs in this specific task. All data and source code will be made publicly available.
☆ Comprehensive Metapath-based Heterogeneous Graph Transformer for Gene-Disease Association Prediction
Discovering gene-disease associations is crucial for understanding disease mechanisms, yet identifying these associations remains challenging due to the time and cost of biological experiments. Computational methods are increasingly vital for efficient and scalable gene-disease association prediction. Graph-based learning models, which leverage node features and network relationships, are commonly employed for biomolecular predictions. However, existing methods often struggle to effectively integrate node features, heterogeneous structures, and semantic information. To address these challenges, we propose COmprehensive MEtapath-based heterogeneous graph Transformer(COMET) for predicting gene-disease associations. COMET integrates diverse datasets to construct comprehensive heterogeneous networks, initializing node features with BioGPT. We define seven Metapaths and utilize a transformer framework to aggregate Metapath instances, capturing global contexts and long-distance dependencies. Through intra- and inter-metapath aggregation using attention mechanisms, COMET fuses latent vectors from multiple Metapaths to enhance GDA prediction accuracy. Our method demonstrates superior robustness compared to state-of-the-art approaches. Ablation studies and visualizations validate COMET's effectiveness, providing valuable insights for advancing human health research.
comment: 6 pages
☆ Derivation of Output Correlation Inferences for Multi-Output (aka Multi-Task) Gaussian Process
Gaussian process (GP) is arguably one of the most widely used machine learning algorithms in practice. One of its prominent applications is Bayesian optimization (BO). Although the vanilla GP itself is already a powerful tool for BO, it is often beneficial to be able to consider the dependencies of multiple outputs. To do so, Multi-task GP (MTGP) is formulated, but it is not trivial to fully understand the derivations of its formulations and their gradients from the previous literature. This paper serves friendly derivations of the MTGP formulations and their gradients.
☆ Self-Instruct Few-Shot Jailbreaking: Decompose the Attack into Pattern and Behavior Learning
Recently, several works have been conducted on jailbreaking Large Language Models (LLMs) with few-shot malicious demos. In particular, Zheng et al. (2024) focuses on improving the efficiency of Few-Shot Jailbreaking (FSJ) by injecting special tokens into the demos and employing demo-level random search. Nevertheless, this method lacks generality since it specifies the instruction-response structure. Moreover, the reason why inserting special tokens takes effect in inducing harmful behaviors is only empirically discussed. In this paper, we take a deeper insight into the mechanism of special token injection and propose Self-Instruct Few-Shot Jailbreaking (Self-Instruct-FSJ) facilitated with the demo-level greedy search. This framework decomposes the FSJ attack into pattern and behavior learning to exploit the model's vulnerabilities in a more generalized and efficient way. We conduct elaborate experiments to evaluate our method on common open-source models and compare it with baseline algorithms. Our code is available at https://github.com/iphosi/Self-Instruct-FSJ.
☆ AI Guide Dog: Egocentric Path Prediction on Smartphone
This paper introduces AI Guide Dog (AIGD), a lightweight egocentric navigation assistance system for visually impaired individuals, designed for real-time deployment on smartphones. AIGD addresses key challenges in blind navigation by employing a vision-only, multi-label classification approach to predict directional commands, ensuring safe traversal across diverse environments. We propose a novel technique to enable goal-based outdoor navigation by integrating GPS signals and high-level directions, while also addressing uncertain multi-path predictions for destination-free indoor navigation. Our generalized model is the first navigation assistance system to handle both goal-oriented and exploratory navigation scenarios across indoor and outdoor settings, establishing a new state-of-the-art in blind navigation. We present methods, datasets, evaluations, and deployment insights to encourage further innovations in assistive navigation systems.
☆ Early prediction of the transferability of bovine embryos from videomicroscopy
Videomicroscopy is a promising tool combined with machine learning for studying the early development of in vitro fertilized bovine embryos and assessing its transferability as soon as possible. We aim to predict the embryo transferability within four days at most, taking 2D time-lapse microscopy videos as input. We formulate this problem as a supervised binary classification problem for the classes transferable and not transferable. The challenges are three-fold: 1) poorly discriminating appearance and motion, 2) class ambiguity, 3) small amount of annotated data. We propose a 3D convolutional neural network involving three pathways, which makes it multi-scale in time and able to handle appearance and motion in different ways. For training, we retain the focal loss. Our model, named SFR, compares favorably to other methods. Experiments demonstrate its effectiveness and accuracy for our challenging biological task.
comment: Accepted at the 2024 IEEE International Conference on Image Processing
☆ Advice for Diabetes Self-Management by ChatGPT Models: Challenges and Recommendations
Given their ability for advanced reasoning, extensive contextual understanding, and robust question-answering abilities, large language models have become prominent in healthcare management research. Despite adeptly handling a broad spectrum of healthcare inquiries, these models face significant challenges in delivering accurate and practical advice for chronic conditions such as diabetes. We evaluate the responses of ChatGPT versions 3.5 and 4 to diabetes patient queries, assessing their depth of medical knowledge and their capacity to deliver personalized, context-specific advice for diabetes self-management. Our findings reveal discrepancies in accuracy and embedded biases, emphasizing the models' limitations in providing tailored advice unless activated by sophisticated prompting techniques. Additionally, we observe that both models often provide advice without seeking necessary clarification, a practice that can result in potentially dangerous advice. This underscores the limited practical effectiveness of these models without human oversight in clinical settings. To address these issues, we propose a commonsense evaluation layer for prompt evaluation and incorporating disease-specific external memory using an advanced Retrieval Augmented Generation technique. This approach aims to improve information quality and reduce misinformation risks, contributing to more reliable AI applications in healthcare settings. Our findings seek to influence the future direction of AI in healthcare, enhancing both the scope and quality of its integration.
☆ An Adaptive Orthogonal Convolution Scheme for Efficient and Flexible CNN Architectures
Orthogonal convolutional layers are the workhorse of multiple areas in machine learning, such as adversarial robustness, normalizing flows, GANs, and Lipschitzconstrained models. Their ability to preserve norms and ensure stable gradient propagation makes them valuable for a large range of problems. Despite their promise, the deployment of orthogonal convolution in large-scale applications is a significant challenge due to computational overhead and limited support for modern features like strides, dilations, group convolutions, and transposed convolutions.In this paper, we introduce AOC (Adaptative Orthogonal Convolution), a scalable method for constructing orthogonal convolutions, effectively overcoming these limitations. This advancement unlocks the construction of architectures that were previously considered impractical. We demonstrate through our experiments that our method produces expressive models that become increasingly efficient as they scale. To foster further advancement, we provide an open-source library implementing this method, available at https://github.com/thib-s/orthogonium.
☆ Gandalf the Red: Adaptive Security for LLMs
Current evaluations of defenses against prompt attacks in large language model (LLM) applications often overlook two critical factors: the dynamic nature of adversarial behavior and the usability penalties imposed on legitimate users by restrictive defenses. We propose D-SEC (Dynamic Security Utility Threat Model), which explicitly separates attackers from legitimate users, models multi-step interactions, and rigorously expresses the security-utility in an optimizable form. We further address the shortcomings in existing evaluations by introducing Gandalf, a crowd-sourced, gamified red-teaming platform designed to generate realistic, adaptive attack datasets. Using Gandalf, we collect and release a dataset of 279k prompt attacks. Complemented by benign user data, our analysis reveals the interplay between security and utility, showing that defenses integrated in the LLM (e.g., system prompts) can degrade usability even without blocking requests. We demonstrate that restricted application domains, defense-in-depth, and adaptive defenses are effective strategies for building secure and useful LLM applications. Code is available at \href{https://github.com/lakeraai/dsec-gandalf}{\texttt{https://github.com/lakeraai/dsec-gandalf}}.
comment: Niklas Pfister, V\'aclav Volhejn and Manuel Knott contributed equally
☆ Exploring Aviation Incident Narratives Using Topic Modeling and Clustering Techniques
Aviation safety is a global concern, requiring detailed investigations into incidents to understand contributing factors comprehensively. This study uses the National Transportation Safety Board (NTSB) dataset. It applies advanced natural language processing (NLP) techniques, including Latent Dirichlet Allocation (LDA), Non-Negative Matrix Factorization (NMF), Latent Semantic Analysis (LSA), Probabilistic Latent Semantic Analysis (pLSA), and K-means clustering. The main objectives are identifying latent themes, exploring semantic relationships, assessing probabilistic connections, and cluster incidents based on shared characteristics. This research contributes to aviation safety by providing insights into incident narratives and demonstrating the versatility of NLP and topic modelling techniques in extracting valuable information from complex datasets. The results, including topics identified from various techniques, provide an understanding of recurring themes. Comparative analysis reveals that LDA performed best with a coherence value of 0.597, pLSA of 0.583, LSA of 0.542, and NMF of 0.437. K-means clustering further reveals commonalities and unique insights into incident narratives. In conclusion, this study uncovers latent patterns and thematic structures within incident narratives, offering a comparative analysis of multiple-topic modelling techniques. Future research avenues include exploring temporal patterns, incorporating additional datasets, and developing predictive models for early identification of safety issues. This research lays the groundwork for enhancing the understanding and improvement of aviation safety by utilising the wealth of information embedded in incident narratives.
☆ Large Language Model Interface for Home Energy Management Systems
Home Energy Management Systems (HEMSs) help households tailor their electricity usage based on power system signals such as energy prices. This technology helps to reduce energy bills and offers greater demand-side flexibility that supports the power system stability. However, residents who lack a technical background may find it difficult to use HEMSs effectively, because HEMSs require well-formatted parameterization that reflects the characteristics of the energy resources, houses, and users' needs. Recently, Large-Language Models (LLMs) have demonstrated an outstanding ability in language understanding. Motivated by this, we propose an LLM-based interface that interacts with users to understand and parameterize their ``badly-formatted answers'', and then outputs well-formatted parameters to implement an HEMS. We further use Reason and Act method (ReAct) and few-shot prompting to enhance the LLM performance. Evaluating the interface performance requires multiple user--LLM interactions. To avoid the efforts in finding volunteer users and reduce the evaluation time, we additionally propose a method that uses another LLM to simulate users with varying expertise, ranging from knowledgeable to non-technical. By comprehensive evaluation, the proposed LLM-based HEMS interface achieves an average parameter retrieval accuracy of 88\%, outperforming benchmark models without ReAct and/or few-shot prompting.
comment: 13 pages conference paper
☆ Governing AI Agents
The field of AI is undergoing a fundamental transition from systems that can produce synthetic content upon request to autonomous agents that can plan and execute complex tasks with only limited human involvement. Companies that pioneered the development of generative AI tools are now building AI agents that can be instructed to independently navigate the internet, perform a wide range of online tasks, and serve as artificial personal assistants and virtual coworkers. The opportunities presented by this new technology are tremendous, as are the associated risks. Fortunately, there exist robust analytic frameworks for confronting many of these challenges, namely, the economic theory of principal-agent problems and the common law doctrine of agency relationships. Drawing on these frameworks, this Article makes three contributions. First, it uses agency law and theory to identify and characterize problems arising from AI agents, including issues of information asymmetry, discretionary authority, and loyalty. Second, it illustrates the limitations of conventional solutions to agency problems: incentive design, monitoring, and enforcement might not be effective for governing AI agents that make uninterpretable decisions and operate at unprecedented speed and scale. Third, the Article explores the implications of agency law and theory for designing and regulating AI agents, arguing that new technical and legal infrastructure is needed to support governance principles of inclusivity, visibility, and liability.
☆ Deep Learning and Natural Language Processing in the Field of Construction
This article presents a complete process to extract hypernym relationships in the field of construction using two main steps: terminology extraction and detection of hypernyms from these terms. We first describe the corpus analysis method to extract terminology from a collection of technical specifications in the field of construction. Using statistics and word n-grams analysis, we extract the domain's terminology and then perform pruning steps with linguistic patterns and internet queries to improve the quality of the final terminology. Second, we present a machine-learning approach based on various words embedding models and combinations to deal with the detection of hypernyms from the extracted terminology. Extracted terminology is evaluated using a manual evaluation carried out by 6 experts in the domain, and the hypernym identification method is evaluated with different datasets. The global approach provides relevant and promising results.
☆ Logarithmic Memory Networks (LMNs): Efficient Long-Range Sequence Modeling for Resource-Constrained Environments
Long-range sequence modeling is a crucial aspect of natural language processing and time series analysis. However, traditional models like Recurrent Neural Networks (RNNs) and Transformers suffer from computational and memory inefficiencies, especially when dealing with long sequences. This paper introduces Logarithmic Memory Networks (LMNs), a novel architecture that leverages a hierarchical logarithmic tree structure to efficiently store and retrieve past information. LMNs dynamically summarize historical context, significantly reducing the memory footprint and computational complexity of attention mechanisms from O(n2) to O(log(n)). The model employs a single-vector, targeted attention mechanism to access stored information, and the memory block construction worker (summarizer) layer operates in two modes: a parallel execution mode during training for efficient processing of hierarchical tree structures and a sequential execution mode during inference, which acts as a memory management system. It also implicitly encodes positional information, eliminating the need for explicit positional encodings. These features make LMNs a robust and scalable solution for processing long-range sequences in resource-constrained environments, offering practical improvements in efficiency and scalability. The code is publicly available under the MIT License on GitHub: https://github.com/AhmedBoin/LogarithmicMemory.
comment: 18 pages, 10 figures
☆ Optimal Classification Trees for Continuous Feature Data Using Dynamic Programming with Branch-and-Bound AAAI-25
Computing an optimal classification tree that provably maximizes training performance within a given size limit, is NP-hard, and in practice, most state-of-the-art methods do not scale beyond computing optimal trees of depth three. Therefore, most methods rely on a coarse binarization of continuous features to maintain scalability. We propose a novel algorithm that optimizes trees directly on the continuous feature data using dynamic programming with branch-and-bound. We develop new pruning techniques that eliminate many sub-optimal splits in the search when similar to previously computed splits and we provide an efficient subroutine for computing optimal depth-two trees. Our experiments demonstrate that these techniques improve runtime by one or more orders of magnitude over state-of-the-art optimal methods and improve test accuracy by 5% over greedy heuristics.
comment: In the proceedings of AAAI-25
☆ Anytime Cooperative Implicit Hitting Set Solving
The Implicit Hitting Set (HS) approach has shown to be very effective for MaxSAT, Pseudo-boolean optimization and other boolean frameworks. Very recently, it has also shown its potential in the very similar Weighted CSP framework by means of the so-called cost-function merging. The original formulation of the HS approach focuses on obtaining increasingly better lower bounds (HS-lb). However, and as shown for Pseudo-Boolean Optimization, this approach can also be adapted to compute increasingly better upper bounds (HS-ub). In this paper we consider both HS approaches and show how they can be easily combined in a multithread architecture where cores discovered by either component are available by the other which, interestingly, generates synergy between them. We show that the resulting algorithm (HS-lub) is consistently superior to either HS-lb and HS-ub in isolation. Most importantly, HS-lub has an effective anytime behaviour with which the optimality gap is reduced during the execution. We tested our approach on the Weighted CSP framework and show on three different benchmarks that our very simple implementation sometimes outperforms the parallel hybrid best-first search implementation of the far more developed state-of-the-art Toulbar2.
☆ Leveraging Metamemory Mechanisms for Enhanced Data-Free Code Generation in LLMs
Automated code generation using large language models (LLMs) has gained attention due to its efficiency and adaptability. However, real-world coding tasks or benchmarks like HumanEval and StudentEval often lack dedicated training datasets, challenging existing few-shot prompting approaches that rely on reference examples. Inspired by human metamemory-a cognitive process involving recall and evaluation-we present a novel framework (namely M^2WF) for improving LLMs' one-time code generation. This approach enables LLMs to autonomously generate, evaluate, and utilize synthetic examples to enhance reliability and performance. Unlike prior methods, it minimizes dependency on curated data and adapts flexibly to various coding scenarios. Our experiments demonstrate significant improvements in coding benchmarks, offering a scalable and robust solution for data-free environments. The code and framework will be publicly available on GitHub and HuggingFace.
comment: 11 pages,6 figures
☆ GRAPHMOE: Amplifying Cognitive Depth of Mixture-of-Experts Network via Introducing Self-Rethinking Mechanism
Traditional Mixture-of-Experts (MoE) networks benefit from utilizing multiple smaller expert models as opposed to a single large network. However, these experts typically operate independently, leaving a question open about whether interconnecting these models could enhance the performance of MoE networks. In response, we introduce GRAPHMOE, a novel method aimed at augmenting the cognitive depth of language models via a self-rethinking mechanism constructed on Pseudo GraphMoE networks. GRAPHMOE employs a recurrent routing strategy to simulate iterative thinking steps, thereby facilitating the flow of information among expert nodes. We implement the GRAPHMOE architecture using Low-Rank Adaptation techniques (LoRA) and conduct extensive experiments on various benchmark datasets. The experimental results reveal that GRAPHMOE outperforms other LoRA based models, achieving state-of-the-art (SOTA) performance. Additionally, this study explores a novel recurrent routing strategy that may inspire further advancements in enhancing the reasoning capabilities of language models.
comment: 10 pages
☆ Tarsier2: Advancing Large Vision-Language Models from Detailed Video Description to Comprehensive Video Understanding
We introduce Tarsier2, a state-of-the-art large vision-language model (LVLM) designed for generating detailed and accurate video descriptions, while also exhibiting superior general video understanding capabilities. Tarsier2 achieves significant advancements through three key upgrades: (1) Scaling pre-training data from 11M to 40M video-text pairs, enriching both volume and diversity; (2) Performing fine-grained temporal alignment during supervised fine-tuning; (3) Using model-based sampling to automatically construct preference data and applying DPO training for optimization. Extensive experiments show that Tarsier2-7B consistently outperforms leading proprietary models, including GPT-4o and Gemini 1.5 Pro, in detailed video description tasks. On the DREAM-1K benchmark, Tarsier2-7B improves F1 by 2.8\% over GPT-4o and 5.8\% over Gemini-1.5-Pro. In human side-by-side evaluations, Tarsier2-7B shows a +8.6\% performance advantage over GPT-4o and +24.9\% over Gemini-1.5-Pro. Tarsier2-7B also sets new state-of-the-art results across 15 public benchmarks, spanning tasks such as video question-answering, video grounding, hallucination test, and embodied question-answering, demonstrating its versatility as a robust generalist vision-language model.
☆ Iterative Label Refinement Matters More than Preference Optimization under Weak Supervision
Language model (LM) post-training relies on two stages of human supervision: task demonstrations for supervised finetuning (SFT), followed by preference comparisons for reinforcement learning from human feedback (RLHF). As LMs become more capable, the tasks they are given become harder to supervise. Will post-training remain effective under unreliable supervision? To test this, we simulate unreliable demonstrations and comparison feedback using small LMs and time-constrained humans. We find that in the presence of unreliable supervision, SFT still retains some effectiveness, but DPO (a common RLHF algorithm) fails to improve the model beyond SFT. To address this, we propose iterative label refinement (ILR) as an alternative to RLHF. ILR improves the SFT data by using comparison feedback to decide whether human demonstrations should be replaced by model-generated alternatives, then retrains the model via SFT on the updated data. SFT+ILR outperforms SFT+DPO on several tasks with unreliable supervision (math, coding, and safe instruction-following). Our findings suggest that as LMs are used for complex tasks where human supervision is unreliable, RLHF may no longer be the best use of human comparison feedback; instead, it is better to direct feedback towards improving the training data rather than continually training the model. Our code and data are available at https://github.com/helloelwin/iterative-label-refinement.
comment: 22 pages, 10 figures
☆ Continual Learning with Embedding Layer Surgery and Task-wise Beam Search using Whisper
Current Multilingual ASR models only support a fraction of the world's languages. Continual Learning (CL) aims to tackle this problem by adding new languages to pre-trained models while avoiding the loss of performance on existing languages, also known as Catastrophic Forgetting (CF). However, existing CL methods overlook the adaptation of the token embedding lookup table at the decoder, despite its significant contribution to CF. We propose Embedding Layer Surgery where separate copies of the token embeddings are created for each new languages, and one of the copies is selected to replace the old languages embeddings when transcribing the corresponding new language. Unfortunately, this approach means LID errors also cause incorrect ASR embedding selection. Our Task-wise Beam Search allows self-correction for such mistakes. By adapting Whisper to 10 hours of data for each of 10 unseen languages from Common Voice, results show that our method reduces the Average WER (AWER) of pre-trained languages from 14.2% to 11.9% compared with Experience Replay, without compromising the AWER of the unseen languages.
comment: Published in 2024 IEEE Spoken Language Technology Workshop
☆ deepTerra -- AI Land Classification Made Easy
deepTerra is a comprehensive platform designed to facilitate the classification of land surface features using machine learning and satellite imagery. The platform includes modules for data collection, image augmentation, training, testing, and prediction, streamlining the entire workflow for image classification tasks. This paper presents a detailed overview of the capabilities of deepTerra, shows how it has been applied to various research areas, and discusses the future directions it might take.
☆ Hierarchical Repository-Level Code Summarization for Business Applications Using Local LLMs ICSE 2025
In large-scale software development, understanding the functionality and intent behind complex codebases is critical for effective development and maintenance. While code summarization has been widely studied, existing methods primarily focus on smaller code units, such as functions, and struggle with larger code artifacts like files and packages. Additionally, current summarization models tend to emphasize low-level implementation details, often overlooking the domain and business context that are crucial for real-world applications. This paper proposes a two-step hierarchical approach for repository-level code summarization, tailored to business applications. First, smaller code units such as functions and variables are identified using syntax analysis and summarized with local LLMs. These summaries are then aggregated to generate higher-level file and package summaries. To ensure the summaries are grounded in business context, we design custom prompts that capture the intended purpose of code artifacts based on the domain and problem context of the business application. We evaluate our approach on a business support system (BSS) for the telecommunications domain, showing that syntax analysis-based hierarchical summarization improves coverage, while business-context grounding enhances the relevance of the generated summaries.
comment: To appear at LLM4Code@ICSE 2025
☆ State-of-the-Art Transformer Models for Image Super-Resolution: Techniques, Challenges, and Applications
Image Super-Resolution (SR) aims to recover a high-resolution image from its low-resolution counterpart, which has been affected by a specific degradation process. This is achieved by enhancing detail and visual quality. Recent advancements in transformer-based methods have remolded image super-resolution by enabling high-quality reconstructions surpassing previous deep-learning approaches like CNN and GAN-based. This effectively addresses the limitations of previous methods, such as limited receptive fields, poor global context capture, and challenges in high-frequency detail recovery. Additionally, the paper reviews recent trends and advancements in transformer-based SR models, exploring various innovative techniques and architectures that combine transformers with traditional networks to balance global and local contexts. These neoteric methods are critically analyzed, revealing promising yet unexplored gaps and potential directions for future research. Several visualizations of models and techniques are included to foster a holistic understanding of recent trends. This work seeks to offer a structured roadmap for researchers at the forefront of deep learning, specifically exploring the impact of transformers on super-resolution techniques.
comment: 8 pages
☆ Optimizing Language Models for Grammatical Acceptability: A Comparative Study of Fine-Tuning Techniques
This study explores the fine-tuning (FT) of the Open Pre-trained Transformer (OPT-125M) for grammatical acceptability tasks using the CoLA dataset. By comparing Vanilla-Fine-Tuning (VFT), Pattern-Based-Fine-Tuning (PBFT), and Parameter-Efficient Fine-Tuning techniques (PEFT) like Low-Rank Adaptation (LoRA), we demonstrate significant improvements in computational efficiency while maintaining high accuracy. Our experiments reveal that while VFT achieves the highest accuracy (81.2%), LoRA enhancing FT by reducing memory usage and iteration time by more than 50%, and increases accuracy in PBFT case. Context Distillation (CD), though computationally efficient, underperformed with accuracy around 31%. Our findings contribute to democratizing access to large language models (LLM) by reducing computational barriers.
☆ Unveiling Provider Bias in Large Language Models for Code Generation
Large Language Models (LLMs) have emerged as the new recommendation engines, outperforming traditional methods in both capability and scope, particularly in code generation applications. Our research reveals a novel provider bias in LLMs, namely without explicit input prompts, these models show systematic preferences for services from specific providers in their recommendations (e.g., favoring Google Cloud over Microsoft Azure). This bias holds significant implications for market dynamics and societal equilibrium, potentially promoting digital monopolies. It may also deceive users and violate their expectations, leading to various consequences. This paper presents the first comprehensive empirical study of provider bias in LLM code generation. We develop a systematic methodology encompassing an automated pipeline for dataset generation, incorporating 6 distinct coding task categories and 30 real-world application scenarios. Our analysis encompasses over 600,000 LLM-generated responses across seven state-of-the-art models, utilizing approximately 500 million tokens (equivalent to \$5,000+ in computational costs). The study evaluates both the generated code snippets and their embedded service provider selections to quantify provider bias. Additionally, we conduct a comparative analysis of seven debiasing prompting techniques to assess their efficacy in mitigating these biases. Our findings demonstrate that LLMs exhibit significant provider preferences, predominantly favoring services from Google and Amazon, and can autonomously modify input code to incorporate their preferred providers without users' requests. Notably, we observe discrepancies between providers recommended in conversational contexts versus those implemented in generated code. The complete dataset and analysis results are available in our repository.
comment: 21 pages, 15 figures
☆ A Driver Advisory System Based on Large Language Model for High-speed Train
With the rapid development of China high-speed railway, drivers face increasingly significant technical challenges during operations, such as fault handling. Currently, drivers depend on the onboard mechanic when facing technical issues, for instance, traction loss or sensor faults. This dependency can hinder effective operation, even lead to accidents, while waiting for faults to be addressed. To enhance the accuracy and explainability of actions during fault handling, an Intelligent Driver Advisory System (IDAS) framework based on a large language model (LLM) named IDAS-LLM, is introduced. Initially, domain-fine-tuning of the LLM is performed using a constructed railway knowledge question-and-answer dataset to improve answer accuracy in railway-related questions. Subsequently, integration of the Retrieval-augmented Generation (RAG) architecture is pursued for system design to enhance the explainability of generated responses. Comparative experiments are conducted using the constructed railway driving knowledge assessment dataset. Results indicate that domain-fine-tuned LLMs show an improvement in answer accuracy by an average of 10%, outperforming some current mainstream LLMs. Additionally, the inclusion of the RAG framework increases the average recall rate of question-and-answer sessions by about 4%. Finally, the fault handling capability of IDAS-LLM is demonstrated through simulations of real operational scenarios, proving that the proposed framework has practical application prospects.
comment: 18 pages, 7 figures, presented at 104th TRB Annual Meeting
☆ Flow: A Modular Approach to Automated Agentic Workflow Generation
Multi-agent frameworks powered by large language models (LLMs) have demonstrated great success in automated planning and task execution. However, the effective adjustment of Agentic workflows during execution has not been well-studied. A effective workflow adjustment is crucial, as in many real-world scenarios, the initial plan must adjust to unforeseen challenges and changing conditions in real-time to ensure the efficient execution of complex tasks. In this paper, we define workflows as an activity-on-vertex (AOV) graphs. We continuously refine the workflow by dynamically adjusting task allocations based on historical performance and previous AOV with LLM agents. To further enhance system performance, we emphasize modularity in workflow design based on measuring parallelism and dependence complexity. Our proposed multi-agent framework achieved efficient sub-task concurrent execution, goal achievement, and error tolerance. Empirical results across different practical tasks demonstrate dramatic improvements in the efficiency of multi-agent frameworks through dynamic workflow updating and modularization.
☆ Real-time Verification and Refinement of Language Model Text Generation
Large language models (LLMs) have shown remarkable performance across a wide range of natural language tasks. However, a critical challenge remains in that they sometimes generate factually incorrect answers. To address this, while many previous work has focused on identifying errors in their generation and further refining them, they are slow in deployment since they are designed to verify the response from LLMs only after their entire generation (from the first to last tokens) is done. Further, we observe that once LLMs generate incorrect tokens early on, there is a higher likelihood that subsequent tokens will also be factually incorrect. To this end, in this work, we propose Streaming-VR (Streaming Verification and Refinement), a novel approach designed to enhance the efficiency of verification and refinement of LLM outputs. Specifically, the proposed Streaming-VR enables on-the-fly verification and correction of tokens as they are being generated, similar to a streaming process, ensuring that each subset of tokens is checked and refined in real-time by another LLM as the LLM constructs its response. Through comprehensive evaluations on multiple datasets, we demonstrate that our approach not only enhances the factual accuracy of LLMs, but also offers a more efficient solution compared to prior refinement methods.
comment: Preprint
☆ A Multi-Encoder Frozen-Decoder Approach for Fine-Tuning Large Language Models
Among parameter-efficient fine-tuning methods, freezing has emerged as a popular strategy for speeding up training, reducing catastrophic forgetting, and improving downstream performance. We investigate the impact of freezing the decoder in a multi-task setup comprising diverse natural language tasks, aiming to reduce deployment overhead and enhance portability to novel tasks. Our experiments, conducted by fine-tuning both individual and multi-task setups on the AlexaTM model, reveal that freezing decoders is highly effective for tasks with natural language outputs and mitigates catastrophic forgetting in multilingual tasks. However, we find that pairing frozen decoders with a larger model can effectively maintain or even enhance performance in structured and QA tasks, making it a viable strategy for a broader range of task types.
☆ Agent-Centric Projection of Prompting Techniques and Implications for Synthetic Training Data for Large Language Models
Recent advances in prompting techniques and multi-agent systems for Large Language Models (LLMs) have produced increasingly complex approaches. However, we lack a framework for characterizing and comparing prompting techniques or understanding their relationship to multi-agent LLM systems. This position paper introduces and explains the concepts of linear contexts (a single, continuous sequence of interactions) and non-linear contexts (branching or multi-path) in LLM systems. These concepts enable the development of an agent-centric projection of prompting techniques, a framework that can reveal deep connections between prompting strategies and multi-agent systems. We propose three conjectures based on this framework: (1) results from non-linear prompting techniques can predict outcomes in equivalent multi-agent systems, (2) multi-agent system architectures can be replicated through single-LLM prompting techniques that simulate equivalent interaction patterns, and (3) these equivalences suggest novel approaches for generating synthetic training data. We argue that this perspective enables systematic cross-pollination of research findings between prompting and multi-agent domains, while providing new directions for improving both the design and training of future LLM systems.
comment: 8 pages, 5 figures. Accepted at ICAART 2025. Derived from an early draft at 2312.17601. arXiv admin note: substantial text overlap with arXiv:2312.17601
☆ STTS-EAD: Improving Spatio-Temporal Learning Based Time Series Prediction via
Handling anomalies is a critical preprocessing step in multivariate time series prediction. However, existing approaches that separate anomaly preprocessing from model training for multivariate time series prediction encounter significant limitations. Specifically, these methods fail to utilize auxiliary information crucial for identifying latent anomalies associated with spatiotemporal factors during the preprocessing stage. Instead, they rely solely on data distribution for anomaly detection, which can result in the incorrect processing of numerous samples that could otherwise contribute positively to model training. To address this, we propose STTS-EAD, an end-to-end method that seamlessly integrates anomaly detection into the training process of multivariate time series forecasting and aims to improve Spatio-Temporal learning based Time Series prediction via Embedded Anomaly Detection. Our proposed STTS-EAD leverages spatio-temporal information for forecasting and anomaly detection, with the two parts alternately executed and optimized for each other. To the best of our knowledge, STTS-EAD is the first to integrate anomaly detection and forecasting tasks in the training phase for improving the accuracy of multivariate time series forecasting. Extensive experiments on a public stock dataset and two real-world sales datasets from a renowned coffee chain enterprise show that our proposed method can effectively process detected anomalies in the training stage to improve forecasting performance in the inference stage and significantly outperform baselines.
comment: 11 pages
☆ Talk to Right Specialists: Routing and Planning in Multi-agent System for Question Answering
Leveraging large language models (LLMs), an agent can utilize retrieval-augmented generation (RAG) techniques to integrate external knowledge and increase the reliability of its responses. Current RAG-based agents integrate single, domain-specific knowledge sources, limiting their ability and leading to hallucinated or inaccurate responses when addressing cross-domain queries. Integrating multiple knowledge bases into a unified RAG-based agent raises significant challenges, including increased retrieval overhead and data sovereignty when sensitive data is involved. In this work, we propose RopMura, a novel multi-agent system that addresses these limitations by incorporating highly efficient routing and planning mechanisms. RopMura features two key components: a router that intelligently selects the most relevant agents based on knowledge boundaries and a planner that decomposes complex multi-hop queries into manageable steps, allowing for coordinating cross-domain responses. Experimental results demonstrate that RopMura effectively handles both single-hop and multi-hop queries, with the routing mechanism enabling precise answers for single-hop queries and the combined routing and planning mechanisms achieving accurate, multi-step resolutions for complex queries.
comment: Work In Progress
☆ Conformal mapping Coordinates Physics-Informed Neural Networks (CoCo-PINNs): learning neural networks for designing neutral inclusions
We focus on designing and solving the neutral inclusion problem via neural networks. The neutral inclusion problem has a long history in the theory of composite materials, and it is exceedingly challenging to identify the precise condition that precipitates a general-shaped inclusion into a neutral inclusion. Physics-informed neural networks (PINNs) have recently become a highly successful approach to addressing both forward and inverse problems associated with partial differential equations. We found that traditional PINNs perform inadequately when applied to the inverse problem of designing neutral inclusions with arbitrary shapes. In this study, we introduce a novel approach, Conformal mapping Coordinates Physics-Informed Neural Networks (CoCo-PINNs), which integrates complex analysis techniques into PINNs. This method exhibits strong performance in solving forward-inverse problems to construct neutral inclusions of arbitrary shapes in two dimensions, where the imperfect interface condition on the inclusion's boundary is modeled by training neural networks. Notably, we mathematically prove that training with a single linear field is sufficient to achieve neutrality for untrained linear fields in arbitrary directions, given a minor assumption. We demonstrate that CoCo-PINNs offer enhanced performances in terms of credibility, consistency, and stability.
☆ A Low-cost and Ultra-lightweight Binary Neural Network for Traffic Signal Recognition
The deployment of neural networks in vehicle platforms and wearable Artificial Intelligence-of-Things (AIOT) scenarios has become a research area that has attracted much attention. With the continuous evolution of deep learning technology, many image classification models are committed to improving recognition accuracy, but this is often accompanied by problems such as large model resource usage, complex structure, and high power consumption, which makes it challenging to deploy on resource-constrained platforms. Herein, we propose an ultra-lightweight binary neural network (BNN) model designed for hardware deployment, and conduct image classification research based on the German Traffic Sign Recognition Benchmark (GTSRB) dataset. In addition, we also verify it on the Chinese Traffic Sign (CTS) and Belgian Traffic Sign (BTS) datasets. The proposed model shows excellent recognition performance with an accuracy of up to 97.64%, making it one of the best performing BNN models in the GTSRB dataset. Compared with the full-precision model, the accuracy loss is controlled within 1%, and the parameter storage overhead of the model is only 10% of that of the full-precision model. More importantly, our network model only relies on logical operations and low-bit width fixed-point addition and subtraction operations during the inference phase, which greatly simplifies the design complexity of the processing element (PE). Our research shows the great potential of BNN in the hardware deployment of computer vision models, especially in the field of computer vision tasks related to autonomous driving.
☆ Visual Language Models as Operator Agents in the Space Domain
This paper explores the application of Vision-Language Models (VLMs) as operator agents in the space domain, focusing on both software and hardware operational paradigms. Building on advances in Large Language Models (LLMs) and their multimodal extensions, we investigate how VLMs can enhance autonomous control and decision-making in space missions. In the software context, we employ VLMs within the Kerbal Space Program Differential Games (KSPDG) simulation environment, enabling the agent to interpret visual screenshots of the graphical user interface to perform complex orbital maneuvers. In the hardware context, we integrate VLMs with robotic systems equipped with cameras to inspect and diagnose physical space objects, such as satellites. Our results demonstrate that VLMs can effectively process visual and textual data to generate contextually appropriate actions, competing with traditional methods and non-multimodal LLMs in simulation tasks, and showing promise in real-world applications.
comment: Updated version of the paper presented in 2025 AIAA SciTech. https://arc.aiaa.org/doi/10.2514/6.2025-1543
☆ A Comparative Analysis of DNN-based White-Box Explainable AI Methods in Network Security
New research focuses on creating artificial intelligence (AI) solutions for network intrusion detection systems (NIDS), drawing its inspiration from the ever-growing number of intrusions on networked systems, increasing its complexity and intelligibility. Hence, the use of explainable AI (XAI) techniques in real-world intrusion detection systems comes from the requirement to comprehend and elucidate black-box AI models to security analysts. In an effort to meet such requirements, this paper focuses on applying and evaluating White-Box XAI techniques (particularly LRP, IG, and DeepLift) for NIDS via an end-to-end framework for neural network models, using three widely used network intrusion datasets (NSL-KDD, CICIDS-2017, and RoEduNet-SIMARGL2021), assessing its global and local scopes, and examining six distinct assessment measures (descriptive accuracy, sparsity, stability, robustness, efficiency, and completeness). We also compare the performance of white-box XAI methods with black-box XAI methods. The results show that using White-box XAI techniques scores high in robustness and completeness, which are crucial metrics for IDS. Moreover, the source codes for the programs developed for our XAI evaluation framework are available to be improved and used by the research community.
☆ BioPose: Biomechanically-accurate 3D Pose Estimation from Monocular Videos
Recent advancements in 3D human pose estimation from single-camera images and videos have relied on parametric models, like SMPL. However, these models oversimplify anatomical structures, limiting their accuracy in capturing true joint locations and movements, which reduces their applicability in biomechanics, healthcare, and robotics. Biomechanically accurate pose estimation, on the other hand, typically requires costly marker-based motion capture systems and optimization techniques in specialized labs. To bridge this gap, we propose BioPose, a novel learning-based framework for predicting biomechanically accurate 3D human pose directly from monocular videos. BioPose includes three key components: a Multi-Query Human Mesh Recovery model (MQ-HMR), a Neural Inverse Kinematics (NeurIK) model, and a 2D-informed pose refinement technique. MQ-HMR leverages a multi-query deformable transformer to extract multi-scale fine-grained image features, enabling precise human mesh recovery. NeurIK treats the mesh vertices as virtual markers, applying a spatial-temporal network to regress biomechanically accurate 3D poses under anatomical constraints. To further improve 3D pose estimations, a 2D-informed refinement step optimizes the query tokens during inference by aligning the 3D structure with 2D pose observations. Experiments on benchmark datasets demonstrate that BioPose significantly outperforms state-of-the-art methods. Project website: \url{https://m-usamasaleem.github.io/publication/BioPose/BioPose.html}.
☆ Transforming Indoor Localization: Advanced Transformer Architecture for NLOS Dominated Wireless Environments with Distributed Sensors
Indoor localization in challenging non-line-of-sight (NLOS) environments often leads to mediocre accuracy with traditional approaches. Deep learning (DL) has been applied to tackle these challenges; however, many DL approaches overlook computational complexity, especially for floating-point operations (FLOPs), making them unsuitable for resource-limited devices. Transformer-based models have achieved remarkable success in natural language processing (NLP) and computer vision (CV) tasks, motivating their use in wireless applications. However, their use in indoor localization remains nascent, and directly applying Transformers for indoor localization can be both computationally intensive and exhibit limitations in accuracy. To address these challenges, in this work, we introduce a novel tokenization approach, referred to as Sensor Snapshot Tokenization (SST), which preserves variable-specific representations of power delay profile (PDP) and enhances attention mechanisms by effectively capturing multi-variate correlation. Complementing this, we propose a lightweight Swish-Gated Linear Unit-based Transformer (L-SwiGLU Transformer) model, designed to reduce computational complexity without compromising localization accuracy. Together, these contributions mitigate the computational burden and dependency on large datasets, making Transformer models more efficient and suitable for resource-constrained scenarios. The proposed tokenization method enables the Vanilla Transformer to achieve a 90th percentile positioning error of 0.388 m in a highly NLOS indoor factory, surpassing conventional tokenization methods. The L-SwiGLU ViT further reduces the error to 0.355 m, achieving an 8.51% improvement. Additionally, the proposed model outperforms a 14.1 times larger model with a 46.13% improvement, underscoring its computational efficiency.
comment: The paper has been submitted to IEEE Transactions on Machine Learning in Communications and Networking
☆ Large Language Models for Knowledge Graph Embedding Techniques, Methods, and Challenges: A Survey
Large Language Models (LLMs) have attracted a lot of attention in various fields due to their superior performance, aiming to train hundreds of millions or more parameters on large amounts of text data to understand and generate natural language. As the superior performance of LLMs becomes apparent, they are increasingly being applied to knowledge graph embedding (KGE) related tasks to improve the processing results. As a deep learning model in the field of Natural Language Processing (NLP), it learns a large amount of textual data to predict the next word or generate content related to a given text. However, LLMs have recently been invoked to varying degrees in different types of KGE related scenarios such as multi-modal KGE and open KGE according to their task characteristics. In this paper, we investigate a wide range of approaches for performing LLMs-related tasks in different types of KGE scenarios. To better compare the various approaches, we summarize each KGE scenario in a classification. In addition to the categorization methods, we provide a tabular overview of the methods and their source code links for a more direct comparison. In the article we also discuss the applications in which the methods are mainly used and suggest several forward-looking directions for the development of this new research area.
☆ Deep Learning for Disease Outbreak Prediction: A Robust Early Warning Signal for Transcritical Bifurcations
Early Warning Signals (EWSs) are vital for implementing preventive measures before a disease turns into a pandemic. While new diseases exhibit unique behaviors, they often share fundamental characteristics from a dynamical systems perspective. Moreover, measurements during disease outbreaks are often corrupted by different noise sources, posing challenges for Time Series Classification (TSC) tasks. In this study, we address the problem of having a robust EWS for disease outbreak prediction using a best-performing deep learning model in the domain of TSC. We employed two simulated datasets to train the model: one representing generated dynamical systems with randomly selected polynomial terms to model new disease behaviors, and another simulating noise-induced disease dynamics to account for noisy measurements. The model's performance was analyzed using both simulated data from different disease models and real-world data, including influenza and COVID-19. Results demonstrate that the proposed model outperforms previous models, effectively providing EWSs of impending outbreaks across various scenarios. This study bridges advancements in deep learning with the ability to provide robust early warning signals in noisy environments, making it highly applicable to real-world crises involving emerging disease outbreaks.
comment: 14 pages, 1 figure, 5 tables
☆ On the Statistical Capacity of Deep Generative Models
Deep generative models are routinely used in generating samples from complex, high-dimensional distributions. Despite their apparent successes, their statistical properties are not well understood. A common assumption is that with enough training data and sufficiently large neural networks, deep generative model samples will have arbitrarily small errors in sampling from any continuous target distribution. We set up a unifying framework that debunks this belief. We demonstrate that broad classes of deep generative models, including variational autoencoders and generative adversarial networks, are not universal generators. Under the predominant case of Gaussian latent variables, these models can only generate concentrated samples that exhibit light tails. Using tools from concentration of measure and convex geometry, we give analogous results for more general log-concave and strongly log-concave latent variable distributions. We extend our results to diffusion models via a reduction argument. We use the Gromov--Levy inequality to give similar guarantees when the latent variables lie on manifolds with positive Ricci curvature. These results shed light on the limited capacity of common deep generative models to handle heavy tails. We illustrate the empirical relevance of our work with simulations and financial data.
☆ PSReg: Prior-guided Sparse Mixture of Experts for Point Cloud Registration AAAI 2025
The discriminative feature is crucial for point cloud registration. Recent methods improve the feature discriminative by distinguishing between non-overlapping and overlapping region points. However, they still face challenges in distinguishing the ambiguous structures in the overlapping regions. Therefore, the ambiguous features they extracted resulted in a significant number of outlier matches from overlapping regions. To solve this problem, we propose a prior-guided SMoE-based registration method to improve the feature distinctiveness by dispatching the potential correspondences to the same experts. Specifically, we propose a prior-guided SMoE module by fusing prior overlap and potential correspondence embeddings for routing, assigning tokens to the most suitable experts for processing. In addition, we propose a registration framework by a specific combination of Transformer layer and prior-guided SMoE module. The proposed method not only pays attention to the importance of locating the overlapping areas of point clouds, but also commits to finding more accurate correspondences in overlapping areas. Our extensive experiments demonstrate the effectiveness of our method, achieving state-of-the-art registration recall (95.7\%/79.3\%) on the 3DMatch/3DLoMatch benchmark. Moreover, we also test the performance on ModelNet40 and demonstrate excellent performance.
comment: Accepted by AAAI 2025
☆ Quantifying the Importance of Data Alignment in Downstream Model Performance
Contrary to the conventional emphasis on dataset size, we explore the role of data alignment -- an often overlooked aspect of data quality -- in training capable Large Language Models (LLMs). To do so, we use the Task2Vec-based alignment coefficient, a quantitative measure of the similarity between two datasets, to quantify the impact of alignment between training data and evaluation data on downstream performance. In particular, we conduct controlled \textit{interventional} experiments for two settings: 1. the impact of increased alignment coefficients between various pre-training (pt) against evaluation datasets, and 2. the impact of increased alignment coefficients between domain specific fine-tuning (ft) against domain specific evaluation. The domain specific task we explore is Autoformalization -- the machine translation task between natural language and code for formal verification. In both settings, we find a strong, predictable negative correlation between the alignment coefficient of a model's training and evaluation data and the model's loss/perplexity on the respective downstream task. These findings suggest a re-evaluation of LLM training approaches, demonstrating the relevance of data alignment compared to data quantity, especially in specialized downstream tasks such as Autoformalization.
☆ Benchmarking Classical, Deep, and Generative Models for Human Activity Recognition
Human Activity Recognition (HAR) has gained significant importance with the growing use of sensor-equipped devices and large datasets. This paper evaluates the performance of three categories of models : classical machine learning, deep learning architectures, and Restricted Boltzmann Machines (RBMs) using five key benchmark datasets of HAR (UCI-HAR, OPPORTUNITY, PAMAP2, WISDM, and Berkeley MHAD). We assess various models, including Decision Trees, Random Forests, Convolutional Neural Networks (CNN), and Deep Belief Networks (DBNs), using metrics such as accuracy, precision, recall, and F1-score for a comprehensive comparison. The results show that CNN models offer superior performance across all datasets, especially on the Berkeley MHAD. Classical models like Random Forest do well on smaller datasets but face challenges with larger, more complex data. RBM-based models also show notable potential, particularly for feature learning. This paper offers a detailed comparison to help researchers choose the most suitable model for HAR tasks.
comment: 48 pages, 21 Figures
☆ Detecting Contextual Anomalies by Discovering Consistent Spatial Regions
We describe a method for modeling spatial context to enable video anomaly detection. The main idea is to discover regions that share similar object-level activities by clustering joint object attributes using Gaussian mixture models. We demonstrate that this straightforward approach, using orders of magnitude fewer parameters than competing models, achieves state-of-the-art performance in the challenging spatial-context-dependent Street Scene dataset. As a side benefit, the high-resolution discovered regions learned by the model also provide explainable normalcy maps for human operators without the need for any pre-trained segmentation model.
☆ Towards Zero-Shot & Explainable Video Description by Reasoning over Graphs of Events in Space and Time
In the current era of Machine Learning, Transformers have become the de facto approach across a variety of domains, such as computer vision and natural language processing. Transformer-based solutions are the backbone of current state-of-the-art methods for language generation, image and video classification, segmentation, action and object recognition, among many others. Interestingly enough, while these state-of-the-art methods produce impressive results in their respective domains, the problem of understanding the relationship between vision and language is still beyond our reach. In this work, we propose a common ground between vision and language based on events in space and time in an explainable and programmatic way, to connect learning-based vision and language state of the art models and provide a solution to the long standing problem of describing videos in natural language. We validate that our algorithmic approach is able to generate coherent, rich and relevant textual descriptions on videos collected from a variety of datasets, using both standard metrics (e.g. Bleu, ROUGE) and the modern LLM-as-a-Jury approach.
☆ Active Sampling for Node Attribute Completion on Graphs
Node attribute, a type of crucial information for graph analysis, may be partially or completely missing for certain nodes in real world applications. Restoring the missing attributes is expected to benefit downstream graph learning. Few attempts have been made on node attribute completion, but a novel framework called Structure-attribute Transformer (SAT) was recently proposed by using a decoupled scheme to leverage structures and attributes. SAT ignores the differences in contributing to the learning schedule and finding a practical way to model the different importance of nodes with observed attributes is challenging. This paper proposes a novel AcTive Sampling algorithm (ATS) to restore missing node attributes. The representativeness and uncertainty of each node's information are first measured based on graph structure, representation similarity and learning bias. To select nodes as train samples in the next optimization step, a weighting scheme controlled by Beta distribution is then introduced to linearly combine the two properties. Extensive experiments on four public benchmark datasets and two downstream tasks have shown the superiority of ATS in node attribute completion.
☆ FARE: A Deep Learning-Based Framework for Radar-based Face Recognition and Out-of-distribution Detection ICASSP 2025
In this work, we propose a novel pipeline for face recognition and out-of-distribution (OOD) detection using short-range FMCW radar. The proposed system utilizes Range-Doppler and micro Range-Doppler Images. The architecture features a primary path (PP) responsible for the classification of in-distribution (ID) faces, complemented by intermediate paths (IPs) dedicated to OOD detection. The network is trained in two stages: first, the PP is trained using triplet loss to optimize ID face classification. In the second stage, the PP is frozen, and the IPs-comprising simple linear autoencoder networks-are trained specifically for OOD detection. Using our dataset generated with a 60 GHz FMCW radar, our method achieves an ID classification accuracy of 99.30% and an OOD detection AUROC of 96.91%.
comment: Accepted at ICASSP 2025
☆ Modeling Discrimination with Causal Abstraction
A person is directly racially discriminated against only if her race caused her worse treatment. This implies that race is an attribute sufficiently separable from other attributes to isolate its causal role. But race is embedded in a nexus of social factors that resist isolated treatment. If race is socially constructed, in what sense can it cause worse treatment? Some propose that the perception of race, rather than race itself, causes worse treatment. Others suggest that since causal models require modularity, i.e. the ability to isolate causal effects, attempts to causally model discrimination are misguided. This paper addresses the problem differently. We introduce a framework for reasoning about discrimination, in which race is a high-level abstraction of lower-level features. In this framework, race can be modeled as itself causing worse treatment. Modularity is ensured by allowing assumptions about social construction to be precisely and explicitly stated, via an alignment between race and its constituents. Such assumptions can then be subjected to normative and empirical challenges, which lead to different views of when discrimination occurs. By distinguishing constitutive and causal relations, the abstraction framework pinpoints disagreements in the current literature on modeling discrimination, while preserving a precise causal account of discrimination.
☆ Causal vs. Anticausal merging of predictors NeurIPS 2024
We study the differences arising from merging predictors in the causal and anticausal directions using the same data. In particular we study the asymmetries that arise in a simple model where we merge the predictors using one binary variable as target and two continuous variables as predictors. We use Causal Maximum Entropy (CMAXENT) as inductive bias to merge the predictors, however, we expect similar differences to hold also when we use other merging methods that take into account asymmetries between cause and effect. We show that if we observe all bivariate distributions, the CMAXENT solution reduces to a logistic regression in the causal direction and Linear Discriminant Analysis (LDA) in the anticausal direction. Furthermore, we study how the decision boundaries of these two solutions differ whenever we observe only some of the bivariate distributions implications for Out-Of-Variable (OOV) generalisation.
comment: Presented at the 38th Conference on Neural Information Processing Systems (NeurIPS 2024)
☆ SEAL: Speaker Error Correction using Acoustic-conditioned Large Language Models ICASSP 2025
Speaker Diarization (SD) is a crucial component of modern end-to-end ASR pipelines. Traditional SD systems, which are typically audio-based and operate independently of ASR, often introduce speaker errors, particularly during speaker transitions and overlapping speech. Recently, language models including fine-tuned large language models (LLMs) have shown to be effective as a second-pass speaker error corrector by leveraging lexical context in the transcribed output. In this work, we introduce a novel acoustic conditioning approach to provide more fine-grained information from the acoustic diarizer to the LLM. We also show that a simpler constrained decoding strategy reduces LLM hallucinations, while avoiding complicated post-processing. Our approach significantly reduces the speaker error rates by 24-43% across Fisher, Callhome, and RT03-CTS datasets, compared to the first-pass Acoustic SD.
comment: Accepted at ICASSP 2025
☆ CVaR-Based Variational Quantum Optimization for User Association in Handoff-Aware Vehicular Networks
Efficient resource allocation is essential for optimizing various tasks in wireless networks, which are usually formulated as generalized assignment problems (GAP). GAP, as a generalized version of the linear sum assignment problem, involves both equality and inequality constraints that add computational challenges. In this work, we present a novel Conditional Value at Risk (CVaR)-based Variational Quantum Eigensolver (VQE) framework to address GAP in vehicular networks (VNets). Our approach leverages a hybrid quantum-classical structure, integrating a tailored cost function that balances both objective and constraint-specific penalties to improve solution quality and stability. Using the CVaR-VQE model, we handle the GAP efficiently by focusing optimization on the lower tail of the solution space, enhancing both convergence and resilience on noisy intermediate-scale quantum (NISQ) devices. We apply this framework to a user-association problem in VNets, where our method achieves 23.5% improvement compared to the deep neural network (DNN) approach.
comment: Accepted in IEEE International Conference on Communications (ICC 2025)
☆ Cross-Modal Transferable Image-to-Video Attack on Video Quality Metrics
Recent studies have revealed that modern image and video quality assessment (IQA/VQA) metrics are vulnerable to adversarial attacks. An attacker can manipulate a video through preprocessing to artificially increase its quality score according to a certain metric, despite no actual improvement in visual quality. Most of the attacks studied in the literature are white-box attacks, while black-box attacks in the context of VQA have received less attention. Moreover, some research indicates a lack of transferability of adversarial examples generated for one model to another when applied to VQA. In this paper, we propose a cross-modal attack method, IC2VQA, aimed at exploring the vulnerabilities of modern VQA models. This approach is motivated by the observation that the low-level feature spaces of images and videos are similar. We investigate the transferability of adversarial perturbations across different modalities; specifically, we analyze how adversarial perturbations generated on a white-box IQA model with an additional CLIP module can effectively target a VQA model. The addition of the CLIP module serves as a valuable aid in increasing transferability, as the CLIP model is known for its effective capture of low-level semantics. Extensive experiments demonstrate that IC2VQA achieves a high success rate in attacking three black-box VQA models. We compare our method with existing black-box attack strategies, highlighting its superiority in terms of attack success within the same number of iterations and levels of attack strength. We believe that the proposed method will contribute to the deeper analysis of robust VQA metrics.
comment: Accepted for VISAPP 2025
☆ BiDepth Multimodal Neural Network: Bidirectional Depth Deep Learning Arcitecture for Spatial-Temporal Prediction
Accurate prediction of spatial-temporal (ST) information in dynamic systems, such as urban mobility and weather patterns, is a crucial yet challenging problem. The complexity stems from the intricate interplay between spatial proximity and temporal relevance, where both long-term trends and short-term fluctuations are present in convoluted patterns. Existing approaches, including traditional statistical methods and conventional neural networks, may provide inaccurate results due to the lack of an effective mechanism that simultaneously incorporates information at variable temporal depths while maintaining spatial context, resulting in a trade-off between comprehensive long-term historical analysis and responsiveness to short-term new information. To bridge this gap, this paper proposes the BiDepth Multimodal Neural Network (BDMNN) with bidirectional depth modulation that enables a comprehensive understanding of both long-term seasonality and short-term fluctuations, adapting to the complex ST context. Case studies with real-world public data demonstrate significant improvements in prediction accuracy, with a 12% reduction in Mean Squared Error for urban traffic prediction and a 15% improvement in rain precipitation forecasting compared to state-of-the-art benchmarks, without demanding extra computational resources.
comment: This paper has been submitted to Applied Intelligence for review
☆ Addressing Quality Challenges in Deep Learning: The Role of MLOps and Domain Knowledge
Deep learning (DL) systems present unique challenges in software engineering, especially concerning quality attributes like correctness and resource efficiency. While DL models achieve exceptional performance in specific tasks, engineering DL-based systems is still essential. The effort, cost, and potential diminishing returns of continual improvements must be carefully evaluated, as software engineers often face the critical decision of when to stop refining a system relative to its quality attributes. This experience paper explores the role of MLOps practices -- such as monitoring and experiment tracking -- in creating transparent and reproducible experimentation environments that enable teams to assess and justify the impact of design decisions on quality attributes. Furthermore, we report on experiences addressing the quality challenges by embedding domain knowledge into the design of a DL model and its integration within a larger system. The findings offer actionable insights into not only the benefits of domain knowledge and MLOps but also the strategic consideration of when to limit further optimizations in DL projects to maximize overall system quality and reliability.
comment: 6 pages, 1 figure, accepted to the 4th International Conference on AI Engineering - Software Engineering for AI (CAIN)
☆ Towards Best Practices for Open Datasets for LLM Training
Many AI companies are training their large language models (LLMs) on data without the permission of the copyright owners. The permissibility of doing so varies by jurisdiction: in countries like the EU and Japan, this is allowed under certain restrictions, while in the United States, the legal landscape is more ambiguous. Regardless of the legal status, concerns from creative producers have led to several high-profile copyright lawsuits, and the threat of litigation is commonly cited as a reason for the recent trend towards minimizing the information shared about training datasets by both corporate and public interest actors. This trend in limiting data information causes harm by hindering transparency, accountability, and innovation in the broader ecosystem by denying researchers, auditors, and impacted individuals access to the information needed to understand AI models. While this could be mitigated by training language models on open access and public domain data, at the time of writing, there are no such models (trained at a meaningful scale) due to the substantial technical and sociological challenges in assembling the necessary corpus. These challenges include incomplete and unreliable metadata, the cost and complexity of digitizing physical records, and the diverse set of legal and technical skills required to ensure relevance and responsibility in a quickly changing landscape. Building towards a future where AI systems can be trained on openly licensed data that is responsibly curated and governed requires collaboration across legal, technical, and policy domains, along with investments in metadata standards, digitization, and fostering a culture of openness.
♻ ☆ A Multi-Modal Approach for Face Anti-Spoofing in Non-Calibrated Systems using Disparity Maps
Face recognition technologies are increasingly used in various applications, yet they are vulnerable to face spoofing attacks. These spoofing attacks often involve unique 3D structures, such as printed papers or mobile device screens. Although stereo-depth cameras can detect such attacks effectively, their high-cost limits their widespread adoption. Conversely, two-sensor systems without extrinsic calibration offer a cost-effective alternative but are unable to calculate depth using stereo techniques. In this work, we propose a method to overcome this challenge by leveraging facial attributes to derive disparity information and estimate relative depth for anti-spoofing purposes, using non-calibrated systems. We introduce a multi-modal anti-spoofing model, coined Disparity Model, that incorporates created disparity maps as a third modality alongside the two original sensor modalities. We demonstrate the effectiveness of the Disparity Model in countering various spoof attacks using a comprehensive dataset collected from the Intel RealSense ID Solution F455. Our method outperformed existing methods in the literature, achieving an Equal Error Rate (EER) of 1.71% and a False Negative Rate (FNR) of 2.77% at a False Positive Rate (FPR) of 1%. These errors are lower by 2.45% and 7.94% than the errors of the best comparison method, respectively. Additionally, we introduce a model ensemble that addresses 3D spoof attacks as well, achieving an EER of 2.04% and an FNR of 3.83% at an FPR of 1%. Overall, our work provides a state-of-the-art solution for the challenging task of anti-spoofing in non-calibrated systems that lack depth information.
♻ ☆ RMem: Restricted Memory Banks Improve Video Object Segmentation CVPR 2024
With recent video object segmentation (VOS) benchmarks evolving to challenging scenarios, we revisit a simple but overlooked strategy: restricting the size of memory banks. This diverges from the prevalent practice of expanding memory banks to accommodate extensive historical information. Our specially designed "memory deciphering" study offers a pivotal insight underpinning such a strategy: expanding memory banks, while seemingly beneficial, actually increases the difficulty for VOS modules to decode relevant features due to the confusion from redundant information. By restricting memory banks to a limited number of essential frames, we achieve a notable improvement in VOS accuracy. This process balances the importance and freshness of frames to maintain an informative memory bank within a bounded capacity. Additionally, restricted memory banks reduce the training-inference discrepancy in memory lengths compared with continuous expansion. This fosters new opportunities in temporal reasoning and enables us to introduce the previously overlooked "temporal positional embedding." Finally, our insights are embodied in "RMem" ("R" for restricted), a simple yet effective VOS modification that excels at challenging VOS scenarios and establishes new state of the art for object state changes (on the VOST dataset) and long videos (on the Long Videos dataset). Our code and demo are available at https://restricted-memory.github.io/.
comment: CVPR 2024, Project Page: https://restricted-memory.github.io/
♻ ☆ CriSPO: Multi-Aspect Critique-Suggestion-guided Automatic Prompt Optimization for Text Generation AAAI-2025
Existing automatic prompt engineering methods are typically designed for discriminative tasks, where new task prompts are iteratively refined with limited feedback from a single metric reflecting a single aspect. However, these approaches are suboptimal for generative tasks, which require more nuanced guidance beyond a single numeric metric to improve the prompt and optimize multiple aspects of the generated text. To address these challenges, we propose a novel multi-aspect Critique-Suggestion-guided automatic Prompt Optimization (CriSPO) approach. CriSPO introduces a critique-suggestion module as its core component. This module spontaneously discovers aspects, and compares generated and reference texts across these aspects, providing specific suggestions for prompt modification. These clear critiques and actionable suggestions guide a receptive optimizer module to make more substantial changes, exploring a broader and more effective search space. To further improve CriSPO with multi-metric optimization, we introduce an Automatic Suffix Tuning (AST) extension to enhance the performance of task prompts across multiple metrics. We evaluate CriSPO on 4 state-of-the-art LLMs across 4 summarization and 5 QA datasets. Extensive experiments show 3-4% ROUGE score improvement on summarization and substantial improvement of various metrics on QA. Code available at https://github.com/amazon-science/crispo
comment: Accepted to AAAI-2025
♻ ☆ Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models
We present Deep Compression Autoencoder (DC-AE), a new family of autoencoder models for accelerating high-resolution diffusion models. Existing autoencoder models have demonstrated impressive results at a moderate spatial compression ratio (e.g., 8x), but fail to maintain satisfactory reconstruction accuracy for high spatial compression ratios (e.g., 64x). We address this challenge by introducing two key techniques: (1) Residual Autoencoding, where we design our models to learn residuals based on the space-to-channel transformed features to alleviate the optimization difficulty of high spatial-compression autoencoders; (2) Decoupled High-Resolution Adaptation, an efficient decoupled three-phases training strategy for mitigating the generalization penalty of high spatial-compression autoencoders. With these designs, we improve the autoencoder's spatial compression ratio up to 128 while maintaining the reconstruction quality. Applying our DC-AE to latent diffusion models, we achieve significant speedup without accuracy drop. For example, on ImageNet 512x512, our DC-AE provides 19.1x inference speedup and 17.9x training speedup on H100 GPU for UViT-H while achieving a better FID, compared with the widely used SD-VAE-f8 autoencoder. Our code is available at https://github.com/mit-han-lab/efficientvit.
comment: Preprint. First two authors contributed equally to this work. Update: add USiT (UViT+SiT sampler) results
♻ ☆ HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models NeurIPS 2024
In order to thrive in hostile and ever-changing natural environments, mammalian brains evolved to store large amounts of knowledge about the world and continually integrate new information while avoiding catastrophic forgetting. Despite the impressive accomplishments, large language models (LLMs), even with retrieval-augmented generation (RAG), still struggle to efficiently and effectively integrate a large amount of new experiences after pre-training. In this work, we introduce HippoRAG, a novel retrieval framework inspired by the hippocampal indexing theory of human long-term memory to enable deeper and more efficient knowledge integration over new experiences. HippoRAG synergistically orchestrates LLMs, knowledge graphs, and the Personalized PageRank algorithm to mimic the different roles of neocortex and hippocampus in human memory. We compare HippoRAG with existing RAG methods on multi-hop question answering and show that our method outperforms the state-of-the-art methods remarkably, by up to 20%. Single-step retrieval with HippoRAG achieves comparable or better performance than iterative retrieval like IRCoT while being 10-30 times cheaper and 6-13 times faster, and integrating HippoRAG into IRCoT brings further substantial gains. Finally, we show that our method can tackle new types of scenarios that are out of reach of existing methods. Code and data are available at https://github.com/OSU-NLP-Group/HippoRAG.
comment: NeurIPS 2024. Code and data: https://github.com/OSU-NLP-Group/HippoRAG
♻ ☆ A Comprehensive Survey of Foundation Models in Medicine
Foundation models (FMs) are large-scale deep learning models that are developed using large datasets and self-supervised learning methods. These models serve as a base for different downstream tasks, including healthcare. FMs have been adopted with great success across various domains within healthcare. Existing healthcare-based surveys have not yet included all of these domains. Therefore, we provide a detailed survey of FMs in healthcare. We focus on the history, learning strategies, flagship models, applications, and challenges of FMs. We explore how FMs such as the BERT and GPT families are reshaping various healthcare domains, including clinical large language models, medical image analysis, and omics. Furthermore, we provide a detailed taxonomy of healthcare applications facilitated by FMs, such as clinical NLP, medical computer vision, graph learning, and other biology-related tasks. Despite the promising opportunities FMs provide, they also have several associated challenges, which are explained in detail. We also outline open research issues and potential lessons learned to provide researchers and practitioners with insights into the capabilities of FMs in healthcare to advance their deployment and mitigate associated risks.
comment: Currently under review in IEEE REVIEWS IN BIOMEDICAL ENGINEERING
♻ ☆ Text-guided Image Restoration and Semantic Enhancement for Text-to-Image Person Retrieval
The goal of Text-to-Image Person Retrieval (TIPR) is to retrieve specific person images according to the given textual descriptions. A primary challenge in this task is bridging the substantial representational gap between visual and textual modalities. The prevailing methods map texts and images into unified embedding space for matching, while the intricate semantic correspondences between texts and images are still not effectively constructed. To address this issue, we propose a novel TIPR framework to build fine-grained interactions and alignment between person images and the corresponding texts. Specifically, via fine-tuning the Contrastive Language-Image Pre-training (CLIP) model, a visual-textual dual encoder is firstly constructed, to preliminarily align the image and text features. Secondly, a Text-guided Image Restoration (TIR) auxiliary task is proposed to map abstract textual entities to specific image regions, improving the alignment between local textual and visual embeddings. Additionally, a cross-modal triplet loss is presented to handle hard samples, and further enhance the model's discriminability for minor differences. Moreover, a pruning-based text data augmentation approach is proposed to enhance focus on essential elements in descriptions, thereby avoiding excessive model attention to less significant information. The experimental results show our proposed method outperforms state-of-the-art methods on three popular benchmark datasets, and the code will be made publicly available at https://github.com/Delong-liu-bupt/SEN.
comment: The paper was withdrawn due to a dispute among the authors regarding the content of the article
♻ ☆ Logic Augmented Generation
Semantic Knowledge Graphs (SKG) face challenges with scalability, flexibility, contextual understanding, and handling unstructured or ambiguous information. However, they offer formal and structured knowledge enabling highly interpretable and reliable results by means of reasoning and querying. Large Language Models (LLMs) overcome those limitations making them suitable in open-ended tasks and unstructured environments. Nevertheless, LLMs are neither interpretable nor reliable. To solve the dichotomy between LLMs and SKGs we envision Logic Augmented Generation (LAG) that combines the benefits of the two worlds. LAG uses LLMs as Reactive Continuous Knowledge Graphs that can generate potentially infinite relations and tacit knowledge on-demand. SKGs are key for injecting a discrete heuristic dimension with clear logical and factual boundaries. We exemplify LAG in two tasks of collective intelligence, i.e., medical diagnostics and climate projections. Understanding the properties and limitations of LAG, which are still mostly unknown, is of utmost importance for enabling a variety of tasks involving tacit knowledge in order to provide interpretable and effective results.
comment: 10 pages, 2 figures
♻ ☆ Relaxed Rotational Equivariance via $G$-Biases in Vision
Group Equivariant Convolution (GConv) can capture rotational equivariance from original data. It assumes uniform and strict rotational equivariance across all features as the transformations under the specific group. However, the presentation or distribution of real-world data rarely conforms to strict rotational equivariance, commonly referred to as Rotational Symmetry-Breaking (RSB) in the system or dataset, making GConv unable to adapt effectively to this phenomenon. Motivated by this, we propose a simple but highly effective method to address this problem, which utilizes a set of learnable biases called $G$-Biases under the group order to break strict group constraints and then achieve a Relaxed Rotational Equivariant Convolution (RREConv). To validate the efficiency of RREConv, we conduct extensive ablation experiments on the discrete rotational group $\mathcal{C}_n$. Experiments demonstrate that the proposed RREConv-based methods achieve excellent performance compared to existing GConv-based methods in both classification and 2D object detection tasks on the natural image datasets.
♻ ☆ WebWalker: Benchmarking LLMs in Web Traversal
Retrieval-augmented generation (RAG) demonstrates remarkable performance across tasks in open-domain question-answering. However, traditional search engines may retrieve shallow content, limiting the ability of LLMs to handle complex, multi-layered information. To address it, we introduce WebWalkerQA, a benchmark designed to assess the ability of LLMs to perform web traversal. It evaluates the capacity of LLMs to traverse a website's subpages to extract high-quality data systematically. We propose WebWalker, which is a multi-agent framework that mimics human-like web navigation through an explore-critic paradigm. Extensive experimental results show that WebWalkerQA is challenging and demonstrates the effectiveness of RAG combined with WebWalker, through the horizontal and vertical integration in real-world scenarios.
♻ ☆ ORFormer: Occlusion-Robust Transformer for Accurate Facial Landmark Detection WACV 2025
Although facial landmark detection (FLD) has gained significant progress, existing FLD methods still suffer from performance drops on partially non-visible faces, such as faces with occlusions or under extreme lighting conditions or poses. To address this issue, we introduce ORFormer, a novel transformer-based method that can detect non-visible regions and recover their missing features from visible parts. Specifically, ORFormer associates each image patch token with one additional learnable token called the messenger token. The messenger token aggregates features from all but its patch. This way, the consensus between a patch and other patches can be assessed by referring to the similarity between its regular and messenger embeddings, enabling non-visible region identification. Our method then recovers occluded patches with features aggregated by the messenger tokens. Leveraging the recovered features, ORFormer compiles high-quality heatmaps for the downstream FLD task. Extensive experiments show that our method generates heatmaps resilient to partial occlusions. By integrating the resultant heatmaps into existing FLD methods, our method performs favorably against the state of the arts on challenging datasets such as WFLW and COFW.
comment: WACV 2025 Project Link: https://ben0919.github.io/ORFormer/
♻ ☆ Inductive Learning of Logical Theories with LLMs: An Expressivity-Graded Analysis
This work presents a novel systematic methodology to analyse the capabilities and limitations of Large Language Models (LLMs) with feedback from a formal inference engine, on logic theory induction. The analysis is complexity-graded w.r.t. rule dependency structure, allowing quantification of specific inference challenges on LLM performance. Integrating LLMs with formal methods is a promising frontier in the Natural Language Processing field, as an important avenue for improving model inference control and explainability. In particular, inductive learning over complex sets of facts and rules, poses unique challenges for current autoregressive models, as they lack explicit symbolic grounding. While they can be complemented by formal systems, the properties delivered by LLMs regarding inductive learning, are not well understood and quantified. Empirical results indicate that the largest LLMs can achieve competitive results against a SOTA Inductive Logic Programming (ILP) system baseline, but also that tracking long predicate relationship chains is a more difficult obstacle than theory complexity for LLMs.
♻ ☆ Are LLMs Good Literature Review Writers? Evaluating the Literature Review Writing Ability of Large Language Models
The literature review is a crucial form of academic writing that involves complex processes of literature collection, organization, and summarization. The emergence of large language models (LLMs) has introduced promising tools to automate these processes. However, their actual capabilities in writing comprehensive literature reviews remain underexplored, such as whether they can generate accurate and reliable references. To address this gap, we propose a framework to assess the literature review writing ability of LLMs automatically. We evaluate the performance of LLMs across three tasks: generating references, writing abstracts, and writing literature reviews. We employ external tools for a multidimensional evaluation, which includes assessing hallucination rates in references, semantic coverage, and factual consistency with human-written context. By analyzing the experimental results, we find that, despite advancements, even the most sophisticated models still cannot avoid generating hallucinated references. Additionally, different models exhibit varying performance in literature review writing across different disciplines.
comment: 12 pages, 5 figures, 5 tables
♻ ☆ Set-based Neural Network Encoding Without Weight Tying
We propose a neural network weight encoding method for network property prediction that utilizes set-to-set and set-to-vector functions to efficiently encode neural network parameters. Our approach is capable of encoding neural networks in a model zoo of mixed architecture and different parameter sizes as opposed to previous approaches that require custom encoding models for different architectures. Furthermore, our \textbf{S}et-based \textbf{N}eural network \textbf{E}ncoder (SNE) takes into consideration the hierarchical computational structure of neural networks. To respect symmetries inherent in network weight space, we utilize Logit Invariance to learn the required minimal invariance properties. Additionally, we introduce a \textit{pad-chunk-encode} pipeline to efficiently encode neural network layers that is adjustable to computational and memory constraints. We also introduce two new tasks for neural network property prediction: cross-dataset and cross-architecture. In cross-dataset property prediction, we evaluate how well property predictors generalize across model zoos trained on different datasets but of the same architecture. In cross-architecture property prediction, we evaluate how well property predictors transfer to model zoos of different architecture not seen during training. We show that SNE outperforms the relevant baselines on standard benchmarks.
comment: 23 pages
♻ ☆ Addressing Hallucinations in Language Models with Knowledge Graph Embeddings as an Additional Modality
In this paper we present an approach to reduce hallucinations in Large Language Models (LLMs) by incorporating Knowledge Graphs (KGs) as an additional modality. Our method involves transforming input text into a set of KG embeddings and using an adapter to integrate these embeddings into the language model space, without relying on external retrieval processes. To facilitate this, we created WikiEntities, a dataset containing over 3 million Wikipedia texts annotated with entities from Wikidata and their corresponding embeddings from PyTorch-BigGraph. This dataset serves as a valuable resource for training Entity Linking models and adapting the described method to various LLMs using specialized adapters. Our method does not require fine-tuning of the language models themselves; instead, we only train the adapter. This ensures that the model's performance on other tasks is not affected. We trained an adapter for the Mistral 7B, LLaMA 2-7B (chat), and LLaMA 3-8B (instruct) models using this dataset and demonstrated that our approach improves performance on the HaluEval, True-False benchmarks and FEVER dataset. The results indicate that incorporating KGs as a new modality can effectively reduce hallucinations and improve the factual accuracy of language models, all without the need for external retrieval.
♻ ☆ Less is More: The Influence of Pruning on the Explainability of CNNs
Over the last century, deep learning models have become the state-of-the-art for solving complex computer vision problems. These modern computer vision models have millions of parameters, which presents two major challenges: (1) the increased computational requirements hamper the deployment in resource-constrained environments, such as mobile or IoT devices, and (2) explaining the complex decisions of such networks to humans is challenging. Network pruning is a technical approach to reduce the complexity of models, where less important parameters are removed. The work presented in this paper investigates whether this reduction in technical complexity also helps with perceived explainability. To do so, we conducted a pre-study and two human-grounded experiments, assessing the effects of different pruning ratios on explainability. Overall, we evaluate four different compression rates (i.e., 2, 4, 8, and 32) with 37 500 tasks on Mechanical Turk. Results indicate that lower compression rates have a positive influence on explainability, while higher compression rates show negative effects. Furthermore, we were able to identify sweet spots that increase both the perceived explainability and the model's performance.
♻ ☆ Spurious Feature Eraser: Stabilizing Test-Time Adaptation for Vision-Language Foundation Model
Vision-language foundation models have exhibited remarkable success across a multitude of downstream tasks due to their scalability on extensive image-text paired data. However, these models also display significant limitations when applied to downstream tasks, such as fine-grained image classification, as a result of ``decision shortcuts'' that hinder their generalization capabilities. In this work, we find that the CLIP model possesses a rich set of features, encompassing both \textit{desired invariant causal features} and \textit{undesired decision shortcuts}. Moreover, the underperformance of CLIP on downstream tasks originates from its inability to effectively utilize pre-trained features in accordance with specific task requirements. To address this challenge, we propose a simple yet effective method, Spurious Feature Eraser (SEraser), to alleviate the decision shortcuts by erasing the spurious features. Specifically, we introduce a test-time prompt tuning paradigm that optimizes a learnable prompt, thereby compelling the model to exploit invariant features while disregarding decision shortcuts during the inference phase. The proposed method effectively alleviates excessive dependence on potentially misleading spurious information. We conduct comparative analysis of the proposed method against various approaches which validates the significant superiority.
♻ ☆ TrIM, Triangular Input Movement Systolic Array for Convolutional Neural Networks: Architecture and Hardware Implementation
Modern hardware architectures for Convolutional Neural Networks (CNNs), other than targeting high performance, aim at dissipating limited energy. Reducing the data movement cost between the computing cores and the memory is a way to mitigate the energy consumption. Systolic arrays are suitable architectures to achieve this objective: they use multiple processing elements that communicate each other to maximize data utilization, based on proper dataflows like the weight stationary and row stationary. Motivated by this, we have proposed TrIM, an innovative dataflow based on a triangular movement of inputs, and capable to reduce the number of memory accesses by one order of magnitude when compared to state-of-the-art systolic arrays. In this paper, we present a TrIM-based hardware architecture for CNNs. As a showcase, the accelerator is implemented onto a Field Programmable Gate Array (FPGA) to execute the VGG-16 and AlexNet CNNs. The architecture achieves a peak throughput of 453.6 Giga Operations per Second, outperforming a state-of-the-art row stationary systolic array up to ~3x in terms of memory accesses, and being up to ~11.9x more energy-efficient than other FPGA accelerators.
comment: This work has been accepted by IEEE TCAS-I for publication
♻ ☆ MiniRAG: Towards Extremely Simple Retrieval-Augmented Generation
The growing demand for efficient and lightweight Retrieval-Augmented Generation (RAG) systems has highlighted significant challenges when deploying Small Language Models (SLMs) in existing RAG frameworks. Current approaches face severe performance degradation due to SLMs' limited semantic understanding and text processing capabilities, creating barriers for widespread adoption in resource-constrained scenarios. To address these fundamental limitations, we present MiniRAG, a novel RAG system designed for extreme simplicity and efficiency. MiniRAG introduces two key technical innovations: (1) a semantic-aware heterogeneous graph indexing mechanism that combines text chunks and named entities in a unified structure, reducing reliance on complex semantic understanding, and (2) a lightweight topology-enhanced retrieval approach that leverages graph structures for efficient knowledge discovery without requiring advanced language capabilities. Our extensive experiments demonstrate that MiniRAG achieves comparable performance to LLM-based methods even when using SLMs while requiring only 25\% of the storage space. Additionally, we contribute a comprehensive benchmark dataset for evaluating lightweight RAG systems under realistic on-device scenarios with complex queries. We fully open-source our implementation and datasets at: https://github.com/HKUDS/MiniRAG.
♻ ☆ Transformers and Large Language Models for Efficient Intrusion Detection Systems: A Comprehensive Survey
With significant advancements in Transformers LLMs, NLP has extended its reach into many research fields due to its enhanced capabilities in text generation and user interaction. One field benefiting greatly from these advancements is cybersecurity. In cybersecurity, many parameters that need to be protected and exchanged between senders and receivers are in the form of text and tabular data, making NLP a valuable tool in enhancing the security measures of communication protocols. This survey paper provides a comprehensive analysis of the utilization of Transformers and LLMs in cyber-threat detection systems. The methodology of paper selection and bibliometric analysis is outlined to establish a rigorous framework for evaluating existing research. The fundamentals of Transformers are discussed, including background information on various cyber-attacks and datasets commonly used in this field. The survey explores the application of Transformers in IDSs, focusing on different architectures such as Attention-based models, LLMs like BERT and GPT, CNN/LSTM-Transformer hybrids, emerging approaches like ViTs, among others. Furthermore, it explores the diverse environments and applications where Transformers and LLMs-based IDS have been implemented, including computer networks, IoT devices, critical infrastructure protection, cloud computing, SDN, as well as in autonomous vehicles. The paper also addresses research challenges and future directions in this area, identifying key issues such as interpretability, scalability, and adaptability to evolving threats, and more. Finally, the conclusion summarizes the findings and highlights the significance of Transformers and LLMs in enhancing cyber-threat detection capabilities, while also outlining potential avenues for further research and development.
comment: arXiv admin note: text overlap with arXiv:2405.04760 by other authors
♻ ☆ GenSafe: A Generalizable Safety Enhancer for Safe Reinforcement Learning Algorithms Based on Reduced Order Markov Decision Process Model
Safe Reinforcement Learning (SRL) aims to realize a safe learning process for Deep Reinforcement Learning (DRL) algorithms by incorporating safety constraints. However, the efficacy of SRL approaches often relies on accurate function approximations, which are notably challenging to achieve in the early learning stages due to data insufficiency. To address this issue, we introduce in this work a novel Generalizable Safety enhancer (GenSafe) that is able to overcome the challenge of data insufficiency and enhance the performance of SRL approaches. Leveraging model order reduction techniques, we first propose an innovative method to construct a Reduced Order Markov Decision Process (ROMDP) as a low-dimensional approximator of the original safety constraints. Then, by solving the reformulated ROMDP-based constraints, GenSafe refines the actions of the agent to increase the possibility of constraint satisfaction. Essentially, GenSafe acts as an additional safety layer for SRL algorithms. We evaluate GenSafe on multiple SRL approaches and benchmark problems. The results demonstrate its capability to improve safety performance, especially in the early learning phases, while maintaining satisfactory task performance. Our proposed GenSafe not only offers a novel measure to augment existing SRL methods but also shows broad compatibility with various SRL algorithms, making it applicable to a wide range of systems and SRL problems.
♻ ☆ Enhanced Masked Image Modeling to Avoid Model Collapse on Multi-modal MRI Datasets
Multi-modal magnetic resonance imaging (MRI) provides information of lesions for computer-aided diagnosis from different views. Deep learning algorithms are suitable for identifying specific anatomical structures, segmenting lesions, and classifying diseases. Manual labels are limited due to the high expense, which hinders further improvement of accuracy. Self-supervised learning, particularly masked image modeling (MIM), has shown promise in utilizing unlabeled data. However, we spot model collapse when applying MIM to multi-modal MRI datasets. The performance of downstream tasks does not see any improvement following the collapsed model. To solve model collapse, we analyze and address it in two types: complete collapse and dimensional collapse. We find complete collapse occurs because the collapsed loss value in multi-modal MRI datasets falls below the normally converged loss value. Based on this, the hybrid mask pattern (HMP) masking strategy is introduced to elevate the collapsed loss above the normally converged loss value and avoid complete collapse. Additionally, we reveal that dimensional collapse stems from insufficient feature uniformity in MIM. We mitigate dimensional collapse by introducing the pyramid barlow twins (PBT) module as an explicit regularization method. Overall, we construct the enhanced MIM (E-MIM) with HMP and PBT module to avoid model collapse multi-modal MRI. Experiments are conducted on three multi-modal MRI datasets to validate the effectiveness of our approach in preventing both types of model collapse. By preventing model collapse, the training of the model becomes more stable, resulting in a decent improvement in performance for segmentation and classification tasks. The code is available at https://github.com/LinxuanHan/E-MIM.
♻ ☆ Private Collaborative Edge Inference via Over-the-Air Computation
We consider collaborative inference at the wireless edge, where each client's model is trained independently on its local dataset. Clients are queried in parallel to make an accurate decision collaboratively. In addition to maximizing the inference accuracy, we also want to ensure the privacy of local models. To this end, we leverage the superposition property of the multiple access channel to implement bandwidth-efficient multi-user inference methods. We propose different methods for ensemble and multi-view classification that exploit over-the-air computation (OAC). We show that these schemes perform better than their orthogonal counterparts with statistically significant differences while using fewer resources and providing privacy guarantees. We also provide experimental results verifying the benefits of the proposed OAC approach to multi-user inference, and perform an ablation study to demonstrate the effectiveness of our design choices. We share the source code of the framework publicly on Github to facilitate further research and reproducibility.
comment: 17 pages, 8 figures. This work extends from our preliminary study presented at the 2022 IEEE International Symposium on Information Theory [1]. arXiv admin note: text overlap with arXiv:2202.03129
♻ ☆ DIDLM: A SLAM Dataset for Difficult Scenarios Featuring Infrared, Depth Cameras, LIDAR, 4D Radar, and Others under Adverse Weather, Low Light Conditions, and Rough Roads
Adverse weather conditions, low-light environments, and bumpy road surfaces pose significant challenges to SLAM in robotic navigation and autonomous driving. Existing datasets in this field predominantly rely on single sensors or combinations of LiDAR, cameras, and IMUs. However, 4D millimeter-wave radar demonstrates robustness in adverse weather, infrared cameras excel in capturing details under low-light conditions, and depth images provide richer spatial information. Multi-sensor fusion methods also show potential for better adaptation to bumpy roads. Despite some SLAM studies incorporating these sensors and conditions, there remains a lack of comprehensive datasets addressing low-light environments and bumpy road conditions, or featuring a sufficiently diverse range of sensor data. In this study, we introduce a multi-sensor dataset covering challenging scenarios such as snowy weather, rainy weather, nighttime conditions, speed bumps, and rough terrains. The dataset includes rarely utilized sensors for extreme conditions, such as 4D millimeter-wave radar, infrared cameras, and depth cameras, alongside 3D LiDAR, RGB cameras, GPS, and IMU. It supports both autonomous driving and ground robot applications and provides reliable GPS/INS ground truth data, covering structured and semi-structured terrains. We evaluated various SLAM algorithms using this dataset, including RGB images, infrared images, depth images, LiDAR, and 4D millimeter-wave radar. The dataset spans a total of 18.5 km, 69 minutes, and approximately 660 GB, offering a valuable resource for advancing SLAM research under complex and extreme conditions. Our dataset is available at https://github.com/GongWeiSheng/DIDLM.
♻ ☆ Evaluation of Artificial Intelligence Methods for Lead Time Prediction in Non-Cycled Areas of Automotive Production
The present study examines the effectiveness of applying Artificial Intelligence methods in an automotive production environment to predict unknown lead times in a non-cycle-controlled production area. Data structures are analyzed to identify contextual features and then preprocessed using one-hot encoding. Methods selection focuses on supervised machine learning techniques. In supervised learning methods, regression and classification methods are evaluated. Continuous regression based on target size distribution is not feasible. Classification methods analysis shows that Ensemble Learning and Support Vector Machines are the most suitable. Preliminary study results indicate that gradient boosting algorithms LightGBM, XGBoost, and CatBoost yield the best results. After further testing and extensive hyperparameter optimization, the final method choice is the LightGBM algorithm. Depending on feature availability and prediction interval granularity, relative prediction accuracies of up to 90% can be achieved. Further tests highlight the importance of periodic retraining of AI models to accurately represent complex production processes using the database. The research demonstrates that AI methods can be effectively applied to highly variable production data, adding business value by providing an additional metric for various control tasks while outperforming current non AI-based systems.
♻ ☆ PastNet: Introducing Physical Inductive Biases for Spatio-temporal Video Prediction
In this paper, we investigate the challenge of spatio-temporal video prediction task, which involves generating future video frames based on historical spatio-temporal observation streams. Existing approaches typically utilize external information such as semantic maps to improve video prediction accuracy, which often neglect the inherent physical knowledge embedded within videos. Worse still, their high computational costs could impede their applications for high-resolution videos. To address these constraints, we introduce a novel framework called \underline{P}hysics-\underline{a}ssisted \underline{S}patio-\underline{t}emporal \underline{Net}work (PastNet) for high-quality video prediction. The core of PastNet lies in incorporating a spectral convolution operator in the Fourier domain, which efficiently introduces inductive biases from the underlying physical laws. Additionally, we employ a memory bank with the estimated intrinsic dimensionality to discretize local features during the processing of complex spatio-temporal signals, thereby reducing computational costs and facilitating efficient high-resolution video prediction. Extensive experiments on various widely-used spatio-temporal video benchmarks demonstrate the effectiveness and efficiency of the proposed PastNet compared with a range of state-of-the-art methods, particularly in high-resolution scenarios.
comment: 11
♻ ☆ MoPE: Mixture of Prompt Experts for Parameter-Efficient and Scalable Multimodal Fusion
Despite the demonstrated parameter efficiency of prompt-based multimodal fusion methods, their limited adaptivity and expressiveness often result in suboptimal performance compared to other tuning approaches. In this paper, we introduce the Mixture of Prompt Experts (MoPE), the first technique designed to overcome these limitations by decomposing standard prompts to capture instance-level features adaptively. Building on this decomposition, MoPE enhances prompt fusion's expressiveness by leveraging multimodal pairing priors to route the most effective prompt for each instance dynamically. Compared to vanilla prompting, our MoPE-based fusion method exhibits greater expressiveness, scaling more effectively with the training data and the overall number of trainable parameters. We also investigate regularization terms for expert routing, which lead to emergent expert specialization with enhanced adaptiveness and interpretablity. Extensive experiments across six multimodal datasets spanning four modalities demonstrate state-of-the-art performance for prompt fusion, matching or even surpassing the performance of fine-tuning while requiring only 0.8% of the trainable parameters. Project homepage: https://github.com/songrise/MoPE
comment: Under Review, Extended version of arxiv:2312.03734
♻ ☆ UTMath: Math Evaluation with Unit Test via Reasoning-to-Coding Thoughts
The evaluation of mathematical reasoning capabilities is essential for advancing Artificial General Intelligence (AGI). While Large Language Models (LLMs) have shown impressive performance in solving mathematical problems, existing benchmarks such as GSM8K and MATH present limitations, including narrow problem definitions with specific numbers and reliance on predetermined rules that hinder accurate assessments of reasoning and generality. This paper introduces the UTMath Benchmark, a robust evaluation framework designed to assess LLMs through extensive unit tests, with a focus on both the accuracy and generality of model responses. It comprises 1,053 cutting-edge problems spanning nine mathematical domains, with an average of 68 test cases per problem. UTMath is highly challenging, with the best-performing model, o1-mini, solving only 32.57\% of the problems, followed by o1-preview at 27.16\%, and GPT-4o at 26.93\%. Furthermore, we present the Reasoning-to-Coding of Thoughts (RCoT) approach, which encourages LLMs to engage in explicit reasoning prior to code generation, thereby facilitating the production of more sophisticated solutions and enhancing overall performance and efficiency. Additionally, we also release the UTMath-Train training dataset (more than 70k samples), to support the community in further exploring mathematical reasoning. Our benchmark can be accessed via the following link: https://github.com/UTMathGroup/UTMath
♻ ☆ To Analyze and Regulate Human-in-the-loop Learning for Congestion Games
In congestion games, selfish users behave myopically to crowd to the shortest paths, and the social planner designs mechanisms to regulate such selfish routing through information or payment incentives. However, such mechanism design requires the knowledge of time-varying traffic conditions and it is the users themselves to learn and report past road experiences to the social planner (e.g., Waze or Google Maps). When congestion games meet mobile crowdsourcing, it is critical to incentivize selfish users to explore non-shortest paths in the best exploitation-exploration trade-off. First, we consider a simple but fundamental parallel routing network with one deterministic path and multiple stochastic paths for users with an average arrival probability $\lambda$. We prove that the current myopic routing policy (widely used in Waze and Google Maps) misses both exploration (when strong hazard belief) and exploitation (when weak hazard belief) as compared to the social optimum. Due to the myopic policy's under-exploration, we prove that the caused price of anarchy (PoA) is larger than \(\frac{1}{1-\rho^{\frac{1}{\lambda}}}\), which can be arbitrarily large as discount factor \(\rho\rightarrow1\). To mitigate such huge efficiency loss, we propose a novel selective information disclosure (SID) mechanism: we only reveal the latest traffic information to users when they intend to over-explore stochastic paths upon arrival, while hiding such information when they want to under-explore. We prove that our mechanism successfully reduces PoA to be less than~\(2\). Besides the parallel routing network, we further extend our mechanism and PoA results to any linear path graphs with multiple intermediate nodes.
comment: arXiv admin note: substantial text overlap with arXiv:2211.14029
♻ ☆ What type of inference is planning?
Multiple types of inference are available for probabilistic graphical models, e.g., marginal, maximum-a-posteriori, and even marginal maximum-a-posteriori. Which one do researchers mean when they talk about "planning as inference"? There is no consistency in the literature, different types are used, and their ability to do planning is further entangled with specific approximations or additional constraints. In this work we use the variational framework to show that, just like all commonly used types of inference correspond to different weightings of the entropy terms in the variational problem, planning corresponds exactly to a different set of weights. This means that all the tricks of variational inference are readily applicable to planning. We develop an analogue of loopy belief propagation that allows us to perform approximate planning in factored-state Markov decisions processes without incurring intractability due to the exponentially large state space. The variational perspective shows that the previous types of inference for planning are only adequate in environments with low stochasticity, and allows us to characterize each type by its own merits, disentangling the type of inference from the additional approximations that its practical use requires. We validate these results empirically on synthetic MDPs and tasks posed in the International Planning Competition.
comment: Camera-ready version update
♻ ☆ ExPO: Explainable Phonetic Trait-Oriented Network for Speaker Verification
In speaker verification, we use computational method to verify if an utterance matches the identity of an enrolled speaker. This task is similar to the manual task of forensic voice comparison, where linguistic analysis is combined with auditory measurements to compare and evaluate voice samples. Despite much success, we have yet to develop a speaker verification system that offers explainable results comparable to those from manual forensic voice comparison. A novel approach, Explainable Phonetic Trait-Oriented (ExPO) network, is proposed in this paper to introduce the speaker's phonetic trait which describes the speaker's characteristics at the phonetic level, resembling what forensic comparison does. ExPO not only generates utterance-level speaker embeddings but also allows for fine-grained analysis and visualization of phonetic traits, offering an explainable speaker verification process. Furthermore, we investigate phonetic traits from within-speaker and between-speaker variation perspectives to determine which trait is most effective for speaker verification, marking an important step towards explainable speaker verification. Our code is available at https://github.com/mmmmayi/ExPO.
comment: Accepted by IEEE Signal Processing Letters
♻ ☆ Snake Learning: A Communication- and Computation-Efficient Distributed Learning Framework for 6G
In the evolution towards 6G, integrating Artificial Intelligence (AI) with advanced network infrastructure emerges as a pivotal strategy for enhancing network intelligence and resource utilization. Existing distributed learning frameworks like Federated Learning and Split Learning often struggle with significant challenges in dynamic network environments including high synchronization demands, costly communication overhead, severe computing resource consumption, and data heterogeneity across network nodes. These obstacles hinder the applications of ubiquitous computing capabilities of 6G networks, especially in light of the trend of escalating model parameters and training data volumes. To address these challenges effectively, this paper introduces ``Snake Learning", a cost-effective distributed learning framework. Specifically, Snake Learning respects the heterogeneity of inter-node computing capability and local data distribution in 6G networks, and sequentially trains the designated part of model layers on individual nodes. This layer-by-layer serpentine update mechanism contributes to significantly reducing the requirements for storage, memory and communication during the model training phase, and demonstrates superior adaptability and efficiency for both classification and fine-tuning tasks across homogeneous and heterogeneous data distributions.
comment: 8 pages, 9 figures
♻ ☆ VBIM-Net: Variational Born Iterative Network for Inverse Scattering Problems
Recently, studies have shown the potential of integrating field-type iterative methods with deep learning (DL) techniques in solving inverse scattering problems (ISPs). In this article, we propose a novel Variational Born Iterative Network, namely, VBIM-Net, to solve the full-wave ISPs with significantly improved structural rationality and inversion quality. The proposed VBIM-Net emulates the alternating updates of the total electric field and the contrast in the variational Born iterative method (VBIM) by multiple layers of subnetworks. We embed the analytical calculation of the contrast variation into each subnetwork, converting the scattered field residual into an approximate contrast variation and then enhancing it by a U-Net, thus avoiding the requirement of matched measurement dimension and grid resolution as in existing approaches. The total field and contrast of each layer's output is supervised in the loss function of VBIM-Net, imposing soft physical constraints on the variables in the subnetworks, which benefits the model's performance.In addition, we design a training scheme with extra noise to enhance the model's stability. Extensive numerical results on synthetic and experimental data both verify the inversion quality, generalization ability, and robustness of the proposed VBIM-Net. This work may provide some new inspiration for the design of efficient field-type DL schemes.
comment: Accepted by IEEE Transactions on Geoscience and Remote Sensing
♻ ☆ FoMo: A Foundation Model for Mobile Traffic Forecasting with Diffusion Model
Mobile traffic forecasting allows operators to anticipate network dynamics and performance in advance, offering substantial potential for enhancing service quality and improving user experience. However, existing models are often task-oriented and are trained with tailored data, which limits their effectiveness in diverse mobile network tasks of Base Station (BS) deployment, resource allocation, energy optimization, etc. and hinders generalization across different urban environments. Foundation models have made remarkable strides across various domains of NLP and CV due to their multi-tasking adaption and zero/few-shot learning capabilities. In this paper, we propose an innovative Foundation model for Mo}bile traffic forecasting (FoMo), aiming to handle diverse forecasting tasks of short/long-term predictions and distribution generation across multiple cities to support network planning and optimization. FoMo combines diffusion models and transformers, where various spatio-temporal masks are proposed to enable FoMo to learn intrinsic features of different tasks, and a contrastive learning strategy is developed to capture the correlations between mobile traffic and urban contexts, thereby improving its transfer learning capability. Extensive experiments on 9 real-world datasets demonstrate that FoMo outperforms current models concerning diverse forecasting tasks and zero/few-shot learning, showcasing a strong universality.
comment: 11 pages, 7 figures
♻ ☆ FLM-101B: An Open LLM and How to Train It with $100K Budget
Large language models (LLMs) are considered important approaches towards foundational machine intelligence, achieving remarkable success in Natural Language Processing and multimodal tasks, among others. However, the carbon footprints and financial costs originating from heavy pre-training computation is a non-negligible issue. Progressive training methods, inspired by the neurogenesis process that grows neural structures, have shown potential to accelerate LLM pre-training. However, the algorithms, implementation, and practices for progressively training LLMs beyond 100B parameters remain underexplored. In this paper, we show that our model, namely FLM-101B, trained with our growth strategy under a budget of \$100K, reaches 80\% of the baselines' performances with only 10\% of their floating-point operations. We believe that further studies on progressive training will benefit the community by cutting down the costs and promoting green AI. The checkpoint of FLM-101B is released at https://huggingface.co/CofeAI/FLM-101B.
♻ ☆ Exploring Gradient Subspaces: Addressing and Overcoming LoRA's Limitations in Federated Fine-Tuning of Large Language Models
Large Language Models (LLMs) have demonstrated remarkable capabilities across various domains, particularly in task generalization for both text and vision data. While fine-tuning these models can significantly enhance their performance on specific downstream tasks, it often requires high-quality data that cannot be shared due to privacy concerns. Federated Learning (FL) offers a promising solution for collaborative training without direct data sharing. However, many parameter-efficient fine-tuning strategies for LLMs in FL, particularly those based on Low-Rank Adaptation (LoRA), face limitations. In this paper, we critically analyze the convergence and performance guarantees of popular FL frameworks utilizing LoRA, highlighting its suboptimal nature due to constrained subspace learning of low-rank matrices. This limitation hinders effective fine-tuning of LLMs in federated settings. Through rigorous analytical and empirical evaluations, we demonstrate that direct weight averaging outperforms LoRA-based strategies, leading to superior performance for fine-tuned models. Our comprehensive comparison unmasks inefficiencies in LoRA approaches and underscores the advantages of direct weight aggregation. We extend our analysis to low-rank gradient-based optimizers, such as GaLore, used during local training steps. Our findings show that GaLore along with direct-weight aggregation is a more effective approach, outperforming federated LoRA methods like FlexLoRA and FFA-LoRA across both text and image modalities. While privacy remains paramount in FL discourse, our focus is on assessing performance outcomes of federated fine-tuned models and evaluating various FL frameworks from both theoretical and empirical perspectives. Our findings advocate reassessing the reliance on LoRA within FL contexts, paving the way for more efficient training methodologies.
♻ ☆ Random Policy Enables In-Context Reinforcement Learning within Trust Horizons
Pretrained foundation models have exhibited extraordinary in-context learning performance, allowing zero-shot generalization to new tasks not encountered during pretraining. In the case of reinforcement learning (RL), in-context RL (ICRL) emerges when pretraining FMs on decision-making problems in an autoregressive-supervised manner. Nevertheless, current state-of-the-art ICRL algorithms, like Algorithm Distillation, Decision Pretrained Transformer and Decision Importance Transformer, impose stringent requirements on the pretraining dataset concerning the source policies, context information, and action labels. Notably, these algorithms either demand optimal policies or require varying degrees of well-trained behavior policies for all pretraining environments. This significantly hinders the application of ICRL to real-world scenarios, where acquiring optimal or well-trained policies for a substantial volume of real-world training environments can be intractable. To overcome this challenge, we introduce a novel approach, termed State-Action Distillation (SAD), that allows to generate an effective pretraining dataset guided solely by random policies. In particular, SAD selects query states and corresponding action labels by distilling outstanding state-action pairs from the entire state and action spaces by using random policies within a trust horizon, and then inherits the classical autoregressive-supervised mechanism during pretraining. To the best of our knowledge, this is the first work that enables effective ICRL under random policies and random contexts. We also establish quantitative analysis of the trustworthiness as well as the performance guarantees of SAD. Moreover, our empirical results across multiple popular ICRL benchmark environments demonstrate that, on average, SAD outperforms the best baseline by 236.3% in the offline evaluation and by 135.2% in the online evaluation.
♻ ☆ What Makes Cryptic Crosswords Challenging for LLMs? COLING 2025
Cryptic crosswords are puzzles that rely on general knowledge and the solver's ability to manipulate language on different levels, dealing with various types of wordplay. Previous research suggests that solving such puzzles is challenging even for modern NLP models, including Large Language Models (LLMs). However, there is little to no research on the reasons for their poor performance on this task. In this paper, we establish the benchmark results for three popular LLMs: Gemma2, LLaMA3 and ChatGPT, showing that their performance on this task is still significantly below that of humans. We also investigate why these models struggle to achieve superior performance. We release our code and introduced datasets at https://github.com/bodasadallah/decrypting-crosswords.
comment: COLING 2025. arXiv admin note: text overlap with arXiv:2403.12094
♻ ☆ GOMA: Proactive Embodied Cooperative Communication via Goal-Oriented Mental Alignment
Verbal communication plays a crucial role in human cooperation, particularly when the partners only have incomplete information about the task, environment, and each other's mental state. In this paper, we propose a novel cooperative communication framework, Goal-Oriented Mental Alignment (GOMA). GOMA formulates verbal communication as a planning problem that minimizes the misalignment between the parts of agents' mental states that are relevant to the goals. This approach enables an embodied assistant to reason about when and how to proactively initialize communication with humans verbally using natural language to help achieve better cooperation. We evaluate our approach against strong baselines in two challenging environments, Overcooked (a multiplayer game) and VirtualHome (a household simulator). Our experimental results demonstrate that large language models struggle with generating meaningful communication that is grounded in the social and physical context. In contrast, our approach can successfully generate concise verbal communication for the embodied assistant to effectively boost the performance of the cooperation as well as human users' perception of the assistant.
comment: 8 pages, 5 figures
♻ ☆ AdaSociety: An Adaptive Environment with Social Structures for Multi-Agent Decision-Making NeurIPS
Traditional interactive environments limit agents' intelligence growth with fixed tasks. Recently, single-agent environments address this by generating new tasks based on agent actions, enhancing task diversity. We consider the decision-making problem in multi-agent settings, where tasks are further influenced by social connections, affecting rewards and information access. However, existing multi-agent environments lack a combination of adaptive physical surroundings and social connections, hindering the learning of intelligent behaviors. To address this, we introduce AdaSociety, a customizable multi-agent environment featuring expanding state and action spaces, alongside explicit and alterable social structures. As agents progress, the environment adaptively generates new tasks with social structures for agents to undertake. In AdaSociety, we develop three mini-games showcasing distinct social structures and tasks. Initial results demonstrate that specific social structures can promote both individual and collective benefits, though current reinforcement learning and LLM-based algorithms show limited effectiveness in leveraging social structures to enhance performance. Overall, AdaSociety serves as a valuable research platform for exploring intelligence in diverse physical and social settings. The code is available at https://github.com/bigai-ai/AdaSociety.
comment: Accepted at NeurIPS D&B 2024
♻ ☆ Mode-conditioned music learning and composition: a spiking neural network inspired by neuroscience and psychology
Musical mode is one of the most critical element that establishes the framework of pitch organization and determines the harmonic relationships. Previous works often use the simplistic and rigid alignment method, and overlook the diversity of modes. However, in contrast to AI models, humans possess cognitive mechanisms for perceiving the various modes and keys. In this paper, we propose a spiking neural network inspired by brain mechanisms and psychological theories to represent musical modes and keys, ultimately generating musical pieces that incorporate tonality features. Specifically, the contributions are detailed as follows: 1) The model is designed with multiple collaborated subsystems inspired by the structures and functions of corresponding brain regions; 2)We incorporate mechanisms for neural circuit evolutionary learning that enable the network to learn and generate mode-related features in music, reflecting the cognitive processes involved in human music perception. 3)The results demonstrate that the proposed model shows a connection framework closely similar to the Krumhansl-Schmuckler model, which is one of the most significant key perception models in the music psychology domain. 4) Experiments show that the model can generate music pieces with characteristics of the given modes and keys. Additionally, the quantitative assessments of generated pieces reveals that the generating music pieces have both tonality characteristics and the melodic adaptability needed to generate diverse and musical content. By combining insights from neuroscience, psychology, and music theory with advanced neural network architectures, our research aims to create a system that not only learns and generates music but also bridges the gap between human cognition and artificial intelligence.
comment: 18 pages, 8 figures
♻ ☆ Radar Signal Recognition through Self-Supervised Learning and Domain Adaptation
Automatic radar signal recognition (RSR) plays a pivotal role in electronic warfare (EW), as accurately classifying radar signals is critical for informing decision-making processes. Recent advances in deep learning have shown significant potential in improving RSR performance in domains with ample annotated data. However, these methods fall short in EW scenarios where annotated RF data are scarce or impractical to obtain. To address these challenges, we introduce a self-supervised learning (SSL) method which utilises masked signal modelling and RF domain adaption to enhance RSR performance in environments with limited RF samples and labels. Specifically, we investigate pre-training masked autoencoders (MAE) on baseband in-phase and quadrature (I/Q) signals from various RF domains and subsequently transfer the learned representation to the radar domain, where annotated data are limited. Empirical results show that our lightweight self-supervised ResNet model with domain adaptation achieves up to a 17.5% improvement in 1-shot classification accuracy when pre-trained on in-domain signals (i.e., radar signals) and up to a 16.31% improvement when pre-trained on out-of-domain signals (i.e., comm signals), compared to its baseline without SSL. We also provide reference results for several MAE designs and pre-training strategies, establishing a new benchmark for few-shot radar signal classification.
comment: 5 pages, 9 figures
♻ ☆ ELDER: Enhancing Lifelong Model Editing with Mixture-of-LoRA AAAI-25
Large language models (LLMs) require model editing to efficiently update specific knowledge within them and avoid factual errors. Most model editing methods are solely designed for single-time use and result in a significant forgetting effect in lifelong editing scenarios, where sequential edits are conducted over time. Previous approaches manage sequential edits by freezing original parameters and discretely allocating new parameters for each knowledge update. However, these methods lack robustness to minor input variations due to the discrete mapping between data and parameters. To overcome this challenge, we propose ELDER, a novel approach to create a continuous association between data and adapters. ELDER integrates multiple LoRAs through a router network and is trained to establish a smooth data-adapter association, thereby enhancing the edit robustness and generalization of semantically equivalent inputs. To ensure inputs containing the same knowledge will be processed by the same LoRAs, we design a novel loss to guide the model link LoRA allocations with edit knowledge. Furthermore, we propose a deferral mechanism to retain the original LLM capabilities post-edit. Extensive experiments on GPT-2 XL and LLaMA2-7B demonstrate that ELDER effectively edits models in the lifelong setting, outperforming eight baselines while exhibiting strong scalability and preserving LLMs' general abilities on downstream tasks. Our code is available at https://github.com/JiaangL/ELDER.
comment: Accepted by AAAI-25
♻ ☆ AI Foundation Models for Wearable Movement Data in Mental Health Research
Pretrained foundation models and transformer architectures have driven the success of large language models (LLMs) and other modern AI breakthroughs. However, similar advancements in health data modeling remain limited due to the need for innovative adaptations. Wearable movement data offers a valuable avenue for exploration, as it's a core feature in nearly all commercial smartwatches, well established in clinical and mental health research, and the sequential nature of the data shares similarities to language. We introduce the Pretrained Actigraphy Transformer (PAT), the first open source foundation model designed for time-series wearable movement data. Leveraging transformer-based architectures and novel techniques, such as patch embeddings, and pretraining on data from 29,307 participants in a national U.S. sample, PAT achieves state-of-the-art performance in several mental health prediction tasks. PAT is also lightweight and easily interpretable, making it a robust tool for mental health research. GitHub: https://github.com/njacobsonlab/Pretrained-Actigraphy-Transformer/
♻ ☆ A Cascaded Dilated Convolution Approach for Mpox Lesion Classification
The global outbreak of the Mpox virus, classified as a Public Health Emergency of International Concern (PHEIC) by the World Health Organization, presents significant diagnostic challenges due to its visual similarity to other skin lesion diseases. Traditional diagnostic methods for Mpox, which rely on clinical symptoms and laboratory tests, are slow and labor intensive. Deep learning-based approaches for skin lesion classification offer a promising alternative. However, developing a model that balances efficiency with accuracy is crucial to ensure reliable and timely diagnosis without compromising performance. This study introduces the Cascaded Atrous Group Attention (CAGA) framework to address these challenges, combining the Cascaded Atrous Attention module and the Cascaded Group Attention mechanism. The Cascaded Atrous Attention module utilizes dilated convolutions and cascades the outputs to enhance multi-scale representation. This is integrated into the Cascaded Group Attention mechanism, which reduces redundancy in Multi-Head Self-Attention. By integrating the Cascaded Atrous Group Attention module with EfficientViT-L1 as the backbone architecture, this approach achieves state-of-the-art performance, reaching an accuracy of 98% on the Mpox Close Skin Image (MCSI) dataset while reducing model parameters by 37.5% compared to the original EfficientViT-L1. The model's robustness is demonstrated through extensive validation on two additional benchmark datasets, where it consistently outperforms existing approaches.
comment: 8 pages, 4 figures, Submitted to Medical Imaging with Deep Learning
♻ ☆ Cost-Effective Robotic Handwriting System with AI Integration
This paper introduces a cost-effective robotic handwriting system designed to replicate human-like handwriting with high precision. Combining a Raspberry Pi Pico microcontroller, 3D-printed components, and a machine learning-based handwriting generation model implemented via TensorFlow, the system converts user-supplied text into realistic stroke trajectories. By leveraging lightweight 3D-printed materials and efficient mechanical designs, the system achieves a total hardware cost of approximately \$56, significantly undercutting commercial alternatives. Experimental evaluations demonstrate handwriting precision within $\pm$0.3 millimeters and a writing speed of approximately 200 mm/min, positioning the system as a viable solution for educational, research, and assistive applications. This study seeks to lower the barriers to personalized handwriting technologies, making them accessible to a broader audience.
comment: This is an updated version of a paper originally presented at the 2024 IEEE Long Island Systems, Applications and Technology Conference (LISAT)
♻ ☆ Can Go AIs be adversarially robust? AAAI 2025
Prior work found that superhuman Go AIs can be defeated by simple adversarial strategies, especially "cyclic" attacks. In this paper, we study whether adding natural countermeasures can achieve robustness in Go, a favorable domain for robustness since it benefits from incredible average-case capability and a narrow, innately adversarial setting. We test three defenses: adversarial training on hand-constructed positions, iterated adversarial training, and changing the network architecture. We find that though some of these defenses protect against previously discovered attacks, none withstand freshly trained adversaries. Furthermore, most of the reliably effective attacks these adversaries discover are different realizations of the same overall class of cyclic attacks. Our results suggest that building robust AI systems is challenging even with extremely superhuman systems in some of the most tractable settings, and highlight two key gaps: efficient generalization of defenses, and diversity in training. For interactive examples of attacks and a link to our codebase, see https://goattack.far.ai.
comment: 63 pages, AAAI 2025
♻ ☆ $\text{Transformer}^2$: Self-adaptive LLMs
Self-adaptive large language models (LLMs) aim to solve the challenges posed by traditional fine-tuning methods, which are often computationally intensive and static in their ability to handle diverse tasks. We introduce $\text{Transformer}^2$, a novel self-adaptation framework that adapts LLMs for unseen tasks in real-time by selectively adjusting only the singular components of their weight matrices. During inference, $\text{Transformer}^2$ employs a two-pass mechanism: first, a dispatch system identifies the task properties, and then task-specific "expert" vectors, trained using reinforcement learning, are dynamically mixed to obtain targeted behavior for the incoming prompt. Our method outperforms ubiquitous approaches such as LoRA, with fewer parameters and greater efficiency. $\text{Transformer}^2$ demonstrates versatility across different LLM architectures and modalities, including vision-language tasks. $\text{Transformer}^2$ represents a significant leap forward, offering a scalable, efficient solution for enhancing the adaptability and task-specific performance of LLMs, paving the way for truly dynamic, self-organizing AI systems.
comment: 18 panges, 11 figures, 9 tables
♻ ☆ Can AI Help with Your Personal Finances?
In recent years, Large Language Models (LLMs) have emerged as a transformative development in artificial intelligence (AI), drawing significant attention from industry and academia. Trained on vast datasets, these sophisticated AI systems exhibit impressive natural language processing and content generation capabilities. This paper explores the potential of LLMs to address key challenges in personal finance, focusing on the United States. We evaluate several leading LLMs, including OpenAI's ChatGPT, Google's Gemini, Anthropic's Claude, and Meta's Llama, to assess their effectiveness in providing accurate financial advice on topics such as mortgages, taxes, loans, and investments. Our findings show that while these models achieve an average accuracy rate of approximately 70%, they also display notable limitations in certain areas. Specifically, LLMs struggle to provide accurate responses for complex financial queries, with performance varying significantly across different topics. Despite these limitations, the analysis reveals notable improvements in newer versions of these models, highlighting their growing utility for individuals and financial advisors. As these AI systems continue to evolve, their potential for advancing AI-driven applications in personal finance becomes increasingly promising.
♻ ☆ Dissecting Query-Key Interaction in Vision Transformers
Self-attention in vision transformers is often thought to perform perceptual grouping where tokens attend to other tokens with similar embeddings, which could correspond to semantically similar features of an object. However, attending to dissimilar tokens can be beneficial by providing contextual information. We propose to analyze the query-key interaction by the singular value decomposition of the interaction matrix (i.e. ${\textbf{W}_q}^\top\textbf{W}_k$). We find that in many ViTs, especially those with classification training objectives, early layers attend more to similar tokens, while late layers show increased attention to dissimilar tokens, providing evidence corresponding to perceptual grouping and contextualization, respectively. Many of these interactions between features represented by singular vectors are interpretable and semantic, such as attention between relevant objects, between parts of an object, or between the foreground and background. This offers a novel perspective on interpreting the attention mechanism, which contributes to understanding how transformer models utilize context and salient features when processing images.
♻ ☆ ACPO: AI-Enabled Compiler Framework
The key to performance optimization of a program is to decide correctly when a certain transformation should be applied by a compiler. This is an ideal opportunity to apply machine-learning models to speed up the tuning process; while this realization has been around since the late 90s, only recent advancements in ML enabled a practical application of ML to compilers as an end-to-end framework. This paper presents ACPO: An AI-Enabled Compiler Framework, a novel framework that provides LLVM with simple and comprehensive tools to benefit from employing ML models for different optimization passes. We first showcase the high-level view, class hierarchy, and functionalities of ACPO and subsequently, demonstrate \taco{a couple of use cases of ACPO by ML-enabling the Loop Unroll and Function Inlining passes used in LLVM's O3. and finally, describe how ACPO can be leveraged to optimize other passes. Experimental results reveal that the ACPO model for Loop Unroll can gain on average 4%, 3%, 5.4%, and 0.2% compared to LLVM's vanilla O3 optimization when deployed on Polybench, Coral-2, CoreMark, and Graph-500, respectively. Furthermore, by including both Function Inlining and Loop Unroll models, ACPO can provide a combined speedup of 4.5% on Polybench and 2.4% on Cbench when compared with LLVM's O3, respectively.
comment: ACPO (12 pages)
♻ ☆ EPIC: Effective Prompting for Imbalanced-Class Data Synthesis in Tabular Data Classification via Large Language Models NeurIPS 2024
Large language models (LLMs) have demonstrated remarkable in-context learning capabilities across diverse applications. In this work, we explore the effectiveness of LLMs for generating realistic synthetic tabular data, identifying key prompt design elements to optimize performance. We introduce EPIC, a novel approach that leverages balanced, grouped data samples and consistent formatting with unique variable mapping to guide LLMs in generating accurate synthetic data across all classes, even for imbalanced datasets. Evaluations on real-world datasets show that EPIC achieves state-of-the-art machine learning classification performance, significantly improving generation efficiency. These findings highlight the effectiveness of EPIC for synthetic tabular data generation, particularly in addressing class imbalance. Our source code for our work is available at: https://seharanul17.github.io/project-synthetic-tabular-llm/
comment: NeurIPS 2024
♻ ☆ Double Equivariance for Inductive Link Prediction for Both New Nodes and New Relation Types
The task of fully inductive link prediction in knowledge graphs has gained significant attention, with various graph neural networks being proposed to address it. This task presents greater challenges than traditional inductive link prediction tasks with only new nodes, as models must be capable of zero-shot generalization to both unseen nodes and unseen relation types in the inference graph. Despite the development of novel models, a unifying theoretical understanding of their success remains elusive, and the limitations of these methods are not well-studied. In this work, we introduce the concept of double permutation-equivariant representations and demonstrate its necessity for effective performance in this task. We show that many existing models, despite their diverse architectural designs, conform to this framework. However, we also identify inherent limitations in double permutation-equivariant representations, which restrict these models's ability to learn effectively on datasets with varying characteristics. Our findings suggest that while double equivariance is necessary for meta-learning across knowledge graphs from different domains, it is not sufficient. There remains a fundamental gap between double permutation-equivariant models and the concept of foundation models designed to learn patterns across all domains.
♻ ☆ Reciprocal Reward Influence Encourages Cooperation From Self-Interested Agents NeurIPS 2024
Cooperation between self-interested individuals is a widespread phenomenon in the natural world, but remains elusive in interactions between artificially intelligent agents. Instead, naive reinforcement learning algorithms typically converge to Pareto-dominated outcomes in even the simplest of social dilemmas. An emerging literature on opponent shaping has demonstrated the ability to reach prosocial outcomes by influencing the learning of other agents. However, such methods differentiate through the learning step of other agents or optimize for meta-game dynamics, which rely on privileged access to opponents' learning algorithms or exponential sample complexity, respectively. To provide a learning rule-agnostic and sample-efficient alternative, we introduce Reciprocators, reinforcement learning agents which are intrinsically motivated to reciprocate the influence of opponents' actions on their returns. This approach seeks to modify other agents' $Q$-values by increasing their return following beneficial actions (with respect to the Reciprocator) and decreasing it after detrimental actions, guiding them towards mutually beneficial actions without directly differentiating through a model of their policy. We show that Reciprocators can be used to promote cooperation in temporally extended social dilemmas during simultaneous learning. Our code is available at https://github.com/johnlyzhou/reciprocator/.
comment: NeurIPS 2024
♻ ☆ Safety Implications of Explainable Artificial Intelligence in End-to-End Autonomous Driving
The end-to-end learning pipeline is gradually creating a paradigm shift in the ongoing development of highly autonomous vehicles, largely due to advances in deep learning, the availability of large-scale training datasets, and improvements in integrated sensor devices. However, a lack of explainability in real-time decisions with contemporary learning methods impedes user trust and attenuates the widespread deployment and commercialization of such vehicles. Moreover, the issue is exacerbated when these cars are involved in or cause traffic accidents. Consequently, explainability in end-to-end autonomous driving is essential to build trust in vehicular automation. With that said, automotive researchers have not yet rigorously explored safety benefits and consequences of explanations in end-to-end autonomous driving. This paper aims to bridge the gaps between these topics and seeks to answer the following research question: What are safety implications of explanations in end-to-end autonomous driving? In this regard, we first revisit established safety and explainability concepts in end-to-end driving. Furthermore, we present three critical case studies and show the pivotal role of explanations in enhancing self-driving safety. Finally, we describe insights from empirical studies and reveal potential value, limitations, and caveats of practical explainable AI methods with respect to their safety assurance in end-to-end driving.
♻ ☆ Cooperative and Asynchronous Transformer-based Mission Planning for Heterogeneous Teams of Mobile Robots
Cooperative mission planning for heterogeneous teams of mobile robots presents a unique set of challenges, particularly when operating under communication constraints and limited computational resources. To address these challenges, we propose the Cooperative and Asynchronous Transformer-based Mission Planning (CATMiP) framework, which leverages multi-agent reinforcement learning (MARL) to coordinate distributed decision making among agents with diverse sensing, motion, and actuation capabilities, operating under sporadic ad hoc communication. A Class-based Macro-Action Decentralized Partially Observable Markov Decision Process (CMacDec-POMDP) is also formulated to effectively model asynchronous decision-making for heterogeneous teams of agents. The framework utilizes an asynchronous centralized training and distributed execution scheme that is developed based on the Multi-Agent Transformer (MAT) architecture. This design allows a single trained model to generalize to larger environments and accommodate varying team sizes and compositions. We evaluate CATMiP in a 2D grid-world simulation environment and compare its performance against planning-based exploration methods. Results demonstrate CATMiP's superior efficiency, scalability, and robustness to communication dropouts, highlighting its potential for real-world heterogeneous mobile robot systems. The code is available at https://github.com/mylad13/CATMiP.
comment: 27 pages, 8 figures, this work has been submitted to Elsevier for possible publication
♻ ☆ Beyond Sight: Finetuning Generalist Robot Policies with Heterogeneous Sensors via Language Grounding
Interacting with the world is a multi-sensory experience: achieving effective general-purpose interaction requires making use of all available modalities -- including vision, touch, and audio -- to fill in gaps from partial observation. For example, when vision is occluded reaching into a bag, a robot should rely on its senses of touch and sound. However, state-of-the-art generalist robot policies are typically trained on large datasets to predict robot actions solely from visual and proprioceptive observations. In this work, we propose FuSe, a novel approach that enables finetuning visuomotor generalist policies on heterogeneous sensor modalities for which large datasets are not readily available by leveraging natural language as a common cross-modal grounding. We combine a multimodal contrastive loss with a sensory-grounded language generation loss to encode high-level semantics. In the context of robot manipulation, we show that FuSe enables performing challenging tasks that require reasoning jointly over modalities such as vision, touch, and sound in a zero-shot setting, such as multimodal prompting, compositional cross-modal prompting, and descriptions of objects it interacts with. We show that the same recipe is applicable to widely different generalist policies, including both diffusion-based generalist policies and large vision-language-action (VLA) models. Extensive experiments in the real world show that FuSeis able to increase success rates by over 20% compared to all considered baselines.
♻ ☆ Unexploited Information Value in Human-AI Collaboration
Humans and AIs are often paired on decision tasks with the expectation of achieving complementary performance -- where the combination of human and AI outperforms either one alone. However, how to improve performance of a human-AI team is often not clear without knowing more about what particular information and strategies each agent employs. In this paper, we propose a model based in statistical decision theory to analyze human-AI collaboration from the perspective of what information could be used to improve a human or AI decision. We demonstrate our model on a deepfake detection task to investigate seven video-level features by their unexploited value of information. We compare the human alone, AI alone and human-AI team and offer insights on how the AI assistance impacts people's usage of the information and what information that the AI exploits well might be useful for improving human decisions.
♻ ☆ Empowering Persian LLMs for Instruction Following: A Novel Dataset and Training Approach
Instruction-tuned large language models have demonstrated remarkable capabilities in following human instructions across various domains. However, their proficiency remains notably deficient in many low-resource languages. To address this challenge, we begin by introducing FarsInstruct a comprehensive instruction dataset designed to enhance the instruction following ability of large language models specifically for the Persian language a significant yet underrepresented language globally. FarsInstruct encompasses a wide range of task types and datasets, each containing a mix of straightforward to complex manual written instructions, as well as translations from the Public Pool of Prompts, ensuring a rich linguistic and cultural representation. Furthermore, we introduce Co-CoLA, a framework designed to enhance the multi-task adaptability of LoRA-tuned models. Through extensive experimental analyses, our study showcases the effectiveness of the FarsInstruct dataset coupled with training by the Co-CoLA framework, in improving the performance of large language models within the Persian context. As of the current writing, FarsInstruct comprises 197 templates across 21 distinct datasets, and we intend to update it consistently, thus augmenting its applicability.
♻ ☆ U-MATH: A University-Level Benchmark for Evaluating Mathematical Skills in LLMs
The current evaluation of mathematical skills in LLMs is limited, as existing benchmarks are either relatively small, primarily focus on elementary and high-school problems, or lack diversity in topics. Additionally, the inclusion of visual elements in tasks remains largely under-explored. To address these gaps, we introduce U-MATH, a novel benchmark of 1,100 unpublished open-ended university-level problems sourced from teaching materials. It is balanced across six core subjects, with 20% of multimodal problems. Given the open-ended nature of U-MATH problems, we employ an LLM to judge the correctness of generated solutions. To this end, we release $\mu$-MATH, a dataset to evaluate the LLMs' capabilities in judging solutions. The evaluation of general domain, math-specific, and multimodal LLMs highlights the challenges presented by U-MATH. Our findings reveal that LLMs achieve a maximum accuracy of only 63% on text-based tasks, with even lower 45% on visual problems. The solution assessment proves challenging for LLMs, with the best LLM judge having an F1-score of 80% on $\mu$-MATH.
♻ ☆ Using Deep Learning to Design High Aspect Ratio Fusion Devices
The design of fusion devices is typically based on computationally expensive simulations. This can be alleviated using high aspect ratio models that employ a reduced number of free parameters, especially in the case of stellarator optimization where non-axisymmetric magnetic fields with a large parameter space are optimized to satisfy certain performance criteria. However, optimization is still required to find configurations with properties such as low elongation, high rotational transform, finite plasma beta, and good fast particle confinement. In this work, we train a machine learning model to construct configurations with favorable confinement properties by finding a solution to the inverse design problem, that is, obtaining a set of model input parameters for given desired properties. Since the solution of the inverse problem is non-unique, a probabilistic approach, based on mixture density networks, is used. It is shown that optimized configurations can be generated reliably using this method.
♻ ☆ Enhancing Performance of Point Cloud Completion Networks with Consistency Loss
Point cloud completion networks are conventionally trained to minimize the disparities between the completed point cloud and the ground-truth counterpart. However, an incomplete object-level point cloud can have multiple valid completion solutions when it is examined in isolation. This one-to-many mapping issue can cause contradictory supervision signals to the network because the loss function may produce different values for identical input-output pairs of the network. In many cases, this issue could adversely affect the network optimization process. In this work, we propose to enhance the conventional learning objective using a novel completion consistency loss to mitigate the one-to-many mapping problem. Specifically, the proposed consistency loss ensure that a point cloud completion network generates a coherent completion solution for incomplete objects originating from the same source point cloud. Experimental results across multiple well-established datasets and benchmarks demonstrated the proposed completion consistency loss have excellent capability to enhance the completion performance of various existing networks without any modification to the design of the networks. The proposed consistency loss enhances the performance of the point completion network without affecting the inference speed, thereby increasing the accuracy of point cloud completion. Notably, a state-of-the-art point completion network trained with the proposed consistency loss can achieve state-of-the-art accuracy on the challenging new MVP dataset. The code and result of experiment various point completion models using proposed consistency loss will be available at: https://github.com/kaist-avelab/ConsistencyLoss .
comment: First version of Paper "Enhancing Performance of Point Cloud Completion Networks with Consistency Loss" by Kevin Tirta Wijaya and Christofel Rio Goenawan. In process submission to Neurocomputing Journal 2024
♻ ☆ UIFV: Data Reconstruction Attack in Vertical Federated Learning
Vertical Federated Learning (VFL) facilitates collaborative machine learning without the need for participants to share raw private data. However, recent studies have revealed privacy risks where adversaries might reconstruct sensitive features through data leakage during the learning process. Although data reconstruction methods based on gradient or model information are somewhat effective, they reveal limitations in VFL application scenarios. This is because these traditional methods heavily rely on specific model structures and/or have strict limitations on application scenarios. To address this, our study introduces the Unified InverNet Framework into VFL, which yields a novel and flexible approach (dubbed UIFV) that leverages intermediate feature data to reconstruct original data, instead of relying on gradients or model details. The intermediate feature data is the feature exchanged by different participants during the inference phase of VFL. Experiments on four datasets demonstrate that our methods significantly outperform state-of-the-art techniques in attack precision. Our work exposes severe privacy vulnerabilities within VFL systems that pose real threats to practical VFL applications and thus confirms the necessity of further enhancing privacy protection in the VFL architecture.
♻ ☆ Learning Discrete Concepts in Latent Hierarchical Models NeurIPS 2024
Learning concepts from natural high-dimensional data (e.g., images) holds potential in building human-aligned and interpretable machine learning models. Despite its encouraging prospect, formalization and theoretical insights into this crucial task are still lacking. In this work, we formalize concepts as discrete latent causal variables that are related via a hierarchical causal model that encodes different abstraction levels of concepts embedded in high-dimensional data (e.g., a dog breed and its eye shapes in natural images). We formulate conditions to facilitate the identification of the proposed causal model, which reveals when learning such concepts from unsupervised data is possible. Our conditions permit complex causal hierarchical structures beyond latent trees and multi-level directed acyclic graphs in prior work and can handle high-dimensional, continuous observed variables, which is well-suited for unstructured data modalities such as images. We substantiate our theoretical claims with synthetic data experiments. Further, we discuss our theory's implications for understanding the underlying mechanisms of latent diffusion models and provide corresponding empirical evidence for our theoretical insights.
comment: NeurIPS 2024
♻ ☆ NeuroBench: A Framework for Benchmarking Neuromorphic Computing Algorithms and Systems
Neuromorphic computing shows promise for advancing computing efficiency and capabilities of AI applications using brain-inspired principles. However, the neuromorphic research field currently lacks standardized benchmarks, making it difficult to accurately measure technological advancements, compare performance with conventional methods, and identify promising future research directions. Prior neuromorphic computing benchmark efforts have not seen widespread adoption due to a lack of inclusive, actionable, and iterative benchmark design and guidelines. To address these shortcomings, we present NeuroBench: a benchmark framework for neuromorphic computing algorithms and systems. NeuroBench is a collaboratively-designed effort from an open community of researchers across industry and academia, aiming to provide a representative structure for standardizing the evaluation of neuromorphic approaches. The NeuroBench framework introduces a common set of tools and systematic methodology for inclusive benchmark measurement, delivering an objective reference framework for quantifying neuromorphic approaches in both hardware-independent (algorithm track) and hardware-dependent (system track) settings. In this article, we outline tasks and guidelines for benchmarks across multiple application domains, and present initial performance baselines across neuromorphic and conventional approaches for both benchmark tracks. NeuroBench is intended to continually expand its benchmarks and features to foster and track the progress made by the research community.
comment: To appear in Nature Neuromorphic Hardware and Computing collection
♻ ☆ Frontier Models are Capable of In-context Scheming
Frontier models are increasingly trained and deployed as autonomous agent. One safety concern is that AI agents might covertly pursue misaligned goals, hiding their true capabilities and objectives - also known as scheming. We study whether models have the capability to scheme in pursuit of a goal that we provide in-context and instruct the model to strongly follow. We evaluate frontier models on a suite of six agentic evaluations where models are instructed to pursue goals and are placed in environments that incentivize scheming. Our results show that o1, Claude 3.5 Sonnet, Claude 3 Opus, Gemini 1.5 Pro, and Llama 3.1 405B all demonstrate in-context scheming capabilities. They recognize scheming as a viable strategy and readily engage in such behavior. For example, models strategically introduce subtle mistakes into their responses, attempt to disable their oversight mechanisms, and even exfiltrate what they believe to be their model weights to external servers. Additionally, this deceptive behavior proves persistent. When o1 has engaged in scheming, it maintains its deception in over 85% of follow-up questions and often remains deceptive in multi-turn interrogations. Analysis of the models' chains-of-thought reveals that models explicitly reason about these deceptive strategies, providing evidence that the scheming behavior is not accidental. Surprisingly, we also find rare instances where models engage in scheming when only given a goal, without being strongly nudged to pursue it. We observe cases where Claude 3.5 Sonnet strategically underperforms in evaluations in pursuit of being helpful, a goal that was acquired during training rather than in-context. Our findings demonstrate that frontier models now possess capabilities for basic in-context scheming, making the potential of AI agents to engage in scheming behavior a concrete rather than theoretical concern.
♻ ☆ On the Geometry of Deep Learning
In this paper, we overview one promising avenue of progress at the mathematical foundation of deep learning: the connection between deep networks and function approximation by affine splines (continuous piecewise linear functions in multiple dimensions). In particular, we will overview work over the past decade on understanding certain geometrical properties of a deep network's affine spline mapping, in particular how it tessellates its input space. As we will see, the affine spline connection and geometrical viewpoint provide a powerful portal through which to view, analyze, and improve the inner workings of a deep network.
comment: Accepted for publication at 'Notices of the American Mathematical Society'
Robotics 29
☆ VINGS-Mono: Visual-Inertial Gaussian Splatting Monocular SLAM in Large Scenes
VINGS-Mono is a monocular (inertial) Gaussian Splatting (GS) SLAM framework designed for large scenes. The framework comprises four main components: VIO Front End, 2D Gaussian Map, NVS Loop Closure, and Dynamic Eraser. In the VIO Front End, RGB frames are processed through dense bundle adjustment and uncertainty estimation to extract scene geometry and poses. Based on this output, the mapping module incrementally constructs and maintains a 2D Gaussian map. Key components of the 2D Gaussian Map include a Sample-based Rasterizer, Score Manager, and Pose Refinement, which collectively improve mapping speed and localization accuracy. This enables the SLAM system to handle large-scale urban environments with up to 50 million Gaussian ellipsoids. To ensure global consistency in large-scale scenes, we design a Loop Closure module, which innovatively leverages the Novel View Synthesis (NVS) capabilities of Gaussian Splatting for loop closure detection and correction of the Gaussian map. Additionally, we propose a Dynamic Eraser to address the inevitable presence of dynamic objects in real-world outdoor scenes. Extensive evaluations in indoor and outdoor environments demonstrate that our approach achieves localization performance on par with Visual-Inertial Odometry while surpassing recent GS/NeRF SLAM methods. It also significantly outperforms all existing methods in terms of mapping and rendering quality. Furthermore, we developed a mobile app and verified that our framework can generate high-quality Gaussian maps in real time using only a smartphone camera and a low-frequency IMU sensor. To the best of our knowledge, VINGS-Mono is the first monocular Gaussian SLAM method capable of operating in outdoor environments and supporting kilometer-scale large scenes.
☆ FDPP: Fine-tune Diffusion Policy with Human Preference
Imitation learning from human demonstrations enables robots to perform complex manipulation tasks and has recently witnessed huge success. However, these techniques often struggle to adapt behavior to new preferences or changes in the environment. To address these limitations, we propose Fine-tuning Diffusion Policy with Human Preference (FDPP). FDPP learns a reward function through preference-based learning. This reward is then used to fine-tune the pre-trained policy with reinforcement learning (RL), resulting in alignment of pre-trained policy with new human preferences while still solving the original task. Our experiments across various robotic tasks and preferences demonstrate that FDPP effectively customizes policy behavior without compromising performance. Additionally, we show that incorporating Kullback-Leibler (KL) regularization during fine-tuning prevents over-fitting and helps maintain the competencies of the initial policy.
☆ Data-driven Spatial Classification using Multi-Arm Bandits for Monitoring with Energy-Constrained Mobile Robots
We consider the spatial classification problem for monitoring using data collected by a coordinated team of mobile robots. Such classification problems arise in several applications including search-and-rescue and precision agriculture. Specifically, we want to classify the regions of a search environment into interesting and uninteresting as quickly as possible using a team of mobile sensors and mobile charging stations. We develop a data-driven strategy that accommodates the noise in sensed data and the limited energy capacity of the sensors, and generates collision-free motion plans for the team. We propose a bi-level approach, where a high-level planner leverages a multi-armed bandit framework to determine the potential regions of interest for the drones to visit next based on the data collected online. Then, a low-level path planner based on integer programming coordinates the paths for the team to visit the target regions subject to the physical constraints. We characterize several theoretical properties of the proposed approach, including anytime guarantees and task completion time. We show the efficacy of our approach in simulation, and further validate these observations in physical experiments using mobile robots.
comment: 8 pages, 6 figures. See https://www.youtube.com/watch?v=gzulpOcVYzg for an overview of the approach along with videos of the hardware experiments
☆ Hybrid Action Based Reinforcement Learning for Multi-Objective Compatible Autonomous Driving
Reinforcement Learning (RL) has shown excellent performance in solving decision-making and control problems of autonomous driving, which is increasingly applied in diverse driving scenarios. However, driving is a multi-attribute problem, leading to challenges in achieving multi-objective compatibility for current RL methods, especially in both policy execution and policy iteration. On the one hand, the common action space structure with single action type limits driving flexibility or results in large behavior fluctuations during policy execution. On the other hand, the multi-attribute weighted single reward function result in the agent's disproportionate attention to certain objectives during policy iterations. To this end, we propose a Multi-objective Ensemble-Critic reinforcement learning method with Hybrid Parametrized Action for multi-objective compatible autonomous driving. Specifically, a parameterized action space is constructed to generate hybrid driving actions, combining both abstract guidance and concrete control commands. A multi-objective critics architecture is constructed considering multiple attribute rewards, to ensure simultaneously focusing on different driving objectives. Additionally, uncertainty-based exploration strategy is introduced to help the agent faster approach viable driving policy. The experimental results in both the simulated traffic environment and the HighD dataset demonstrate that our method can achieve multi-objective compatible autonomous driving in terms of driving efficiency, action consistency, and safety. It enhances the general performance of the driving while significantly increasing training efficiency.
comment: 12 pages, 9 figures, 5 tables
☆ HydroelasticTouch: Simulation of Tactile Sensors with Hydroelastic Contact Surfaces
Thanks to recent advancements in the development of inexpensive, high-resolution tactile sensors, touch sensing has become popular in contact-rich robotic manipulation tasks. With the surge of data-driven methods and their requirement for substantial datasets, several methods of simulating tactile sensors have emerged in the tactile research community to overcome real-world data collection limitations. These simulation approaches can be split into two main categories: fast but inaccurate (soft) point-contact models and slow but accurate finite element modeling. In this work, we present a novel approach to simulating pressure-based tactile sensors using the hydroelastic contact model, which provides a high degree of physical realism at a reasonable computational cost. This model produces smooth contact forces for soft-to-soft and soft-to-rigid contacts along even non-convex contact surfaces. Pressure values are approximated at each point of the contact surface and can be integrated to calculate sensor outputs. We validate our models' capacity to synthesize real-world tactile data by conducting zero-shot sim-to-real transfer of a model for object state estimation. Our simulation is available as a plug-in to our open-source, MuJoCo-based simulator.
☆ CHEQ-ing the Box: Safe Variable Impedance Learning for Robotic Polishing
Robotic systems are increasingly employed for industrial automation, with contact-rich tasks like polishing requiring dexterity and compliant behaviour. These tasks are difficult to model, making classical control challenging. Deep reinforcement learning (RL) offers a promising solution by enabling the learning of models and control policies directly from data. However, its application to real-world problems is limited by data inefficiency and unsafe exploration. Adaptive hybrid RL methods blend classical control and RL adaptively, combining the strengths of both: structure from control and learning from RL. This has led to improvements in data efficiency and exploration safety. However, their potential for hardware applications remains underexplored, with no evaluations on physical systems to date. Such evaluations are critical to fully assess the practicality and effectiveness of these methods in real-world settings. This work presents an experimental demonstration of the hybrid RL algorithm CHEQ for robotic polishing with variable impedance, a task requiring precise force and velocity tracking. In simulation, we show that variable impedance enhances polishing performance. We compare standalone RL with adaptive hybrid RL, demonstrating that CHEQ achieves effective learning while adhering to safety constraints. On hardware, CHEQ achieves effective polishing behaviour, requiring only eight hours of training and incurring just five failures. These results highlight the potential of adaptive hybrid RL for real-world, contact-rich tasks trained directly on hardware.
☆ AI Guide Dog: Egocentric Path Prediction on Smartphone
This paper introduces AI Guide Dog (AIGD), a lightweight egocentric navigation assistance system for visually impaired individuals, designed for real-time deployment on smartphones. AIGD addresses key challenges in blind navigation by employing a vision-only, multi-label classification approach to predict directional commands, ensuring safe traversal across diverse environments. We propose a novel technique to enable goal-based outdoor navigation by integrating GPS signals and high-level directions, while also addressing uncertain multi-path predictions for destination-free indoor navigation. Our generalized model is the first navigation assistance system to handle both goal-oriented and exploratory navigation scenarios across indoor and outdoor settings, establishing a new state-of-the-art in blind navigation. We present methods, datasets, evaluations, and deployment insights to encourage further innovations in assistive navigation systems.
☆ Low-Contact Grasping of Soft Tissue with Complex Geometry using a Vortex Gripper
Soft tissue manipulation is an integral aspect of most surgical procedures; however, the vast majority of surgical graspers used today are made of hard materials, such as metals or hard plastics. Furthermore, these graspers predominately function by pinching tissue between two hard objects as a method for tissue manipulation. As such, the potential to apply too much force during contact, and thus damage tissue, is inherently high. As an alternative approach, gaspers developed using a pneumatic vortex could potentially levitate soft tissue, enabling manipulation with low or even no contact force. In this paper, we present the design and well as a full factorial study of the force characteristics of the vortex gripper grasping soft surfaces with four common shapes, with convex and concave curvature, and ranging over 10 different radii of curvature, for a total of 40 unique surfaces. By changing the parameters of the nozzle elements in the design of the gripper, it was possible to investigate the influence of the mass flow parameters of the vortex gripper on the lifting force for all of these different soft surfaces. An $\pmb{ex}$ $\pmb{vivo}$ experiment was conducted on grasping biological tissues and soft balls of various shapes to show the advantages and disadvantages of the proposed technology. The obtained results allowed us to find limitations in the use of vortex technology and the following stages of its improvement for medical use.
comment: Submitted to T-MRB
☆ Electrostatic Clutches Enable High-Force Mechanical Multiplexing: Demonstrating Single-Motor Full-Actuation of a 4-DoF Hand
This paper introduces a novel mechanical multiplexing system powered by electrostatic capstan clutches, enabling high-force, single-motor control of multiple degrees of freedom (DoF). The system is capable of both bidirectional single-input single-output time-division and single-input multiple-output multiplexing to actuate a commercial 4-DoF robotic hand with a single motor. Our mechanical multiplexer is also capable of powerless position holding owing to its use of a leadscrew nut acting as the output. Experimental results demonstrate the effectiveness of this approach, achieving individual and simultaneous actuation. This innovation offers a scalable solution for high-DoF robotic systems, providing a path to efficient actuation in robotic platforms.
☆ Toward Zero-Shot User Intent Recognition in Shared Autonomy
A fundamental challenge of shared autonomy is to use high-DoF robots to assist, rather than hinder, humans by first inferring user intent and then empowering the user to achieve their intent. Although successful, prior methods either rely heavily on a priori knowledge of all possible human intents or require many demonstrations and interactions with the human to learn these intents before being able to assist the user. We propose and study a zero-shot, vision-only shared autonomy (VOSA) framework designed to allow robots to use end-effector vision to estimate zero-shot human intents in conjunction with blended control to help humans accomplish manipulation tasks with unknown and dynamically changing object locations. To demonstrate the effectiveness of our VOSA framework, we instantiate a simple version of VOSA on a Kinova Gen3 manipulator and evaluate our system by conducting a user study on three tabletop manipulation tasks. The performance of VOSA matches that of an oracle baseline model that receives privileged knowledge of possible human intents while also requiring significantly less effort than unassisted teleoperation. In more realistic settings, where the set of possible human intents is fully or partially unknown, we demonstrate that VOSA requires less human effort and time than baseline approaches while being preferred by a majority of the participants. Our results demonstrate the efficacy and efficiency of using off-the-shelf vision algorithms to enable flexible and beneficial shared control of a robot manipulator. Code and videos available here: https://sites.google.com/view/zeroshot-sharedautonomy/home.
comment: 10 pages, 6 figures, Accepted to IEEE/ACM International Conference on Human-Robot Interaction (HRI), 2025. Equal Contribution from the first three authors
♻ ☆ FaVoR: Features via Voxel Rendering for Camera Relocalization WACV
Camera relocalization methods range from dense image alignment to direct camera pose regression from a query image. Among these, sparse feature matching stands out as an efficient, versatile, and generally lightweight approach with numerous applications. However, feature-based methods often struggle with significant viewpoint and appearance changes, leading to matching failures and inaccurate pose estimates. To overcome this limitation, we propose a novel approach that leverages a globally sparse yet locally dense 3D representation of 2D features. By tracking and triangulating landmarks over a sequence of frames, we construct a sparse voxel map optimized to render image patch descriptors observed during tracking. Given an initial pose estimate, we first synthesize descriptors from the voxels using volumetric rendering and then perform feature matching to estimate the camera pose. This methodology enables the generation of descriptors for unseen views, enhancing robustness to view changes. We extensively evaluate our method on the 7-Scenes and Cambridge Landmarks datasets. Our results show that our method significantly outperforms existing state-of-the-art feature representation techniques in indoor environments, achieving up to a 39% improvement in median translation error. Additionally, our approach yields comparable results to other methods for outdoor scenarios while maintaining lower memory and computational costs.
comment: Accepted to the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Tucson, Arizona, US, Feb 28-Mar 4, 2025
♻ ☆ Vid2Sim: Realistic and Interactive Simulation from Video for Urban Navigation
Sim-to-real gap has long posed a significant challenge for robot learning in simulation, preventing the deployment of learned models in the real world. Previous work has primarily focused on domain randomization and system identification to mitigate this gap. However, these methods are often limited by the inherent constraints of the simulation and graphics engines. In this work, we propose Vid2Sim, a novel framework that effectively bridges the sim2real gap through a scalable and cost-efficient real2sim pipeline for neural 3D scene reconstruction and simulation. Given a monocular video as input, Vid2Sim can generate photorealistic and physically interactable 3D simulation environments to enable the reinforcement learning of visual navigation agents in complex urban environments. Extensive experiments demonstrate that Vid2Sim significantly improves the performance of urban navigation in the digital twins and real world by 31.2% and 68.3% in success rate compared with agents trained with prior simulation methods.
comment: Project page: https://metadriverse.github.io/vid2sim/
♻ ☆ Virtual Reflections on a Dynamic 2D Eye Model Improve Spatial Reference Identification
The visible orientation of human eyes creates some transparency about people's spatial attention and other mental states. This leads to a dual role for the eyes as a means of sensing and communication. Accordingly, artificial eye models are being explored as communication media in human-machine interaction scenarios. One challenge in the use of eye models for communication consists of resolving spatial reference ambiguities, especially for screen-based models. Here, we introduce an approach for overcoming this challenge through the introduction of reflection-like features that are contingent on artificial eye movements. We conducted a user study with 30 participants in which participants had to use spatial references provided by dynamic eye models to advance in a fast-paced group interaction task. Compared to a non-reflective eye model and a pure reflection mode, their combination in the new approach resulted in a higher identification accuracy and user experience, suggesting a synergistic benefit.
comment: This work has been submitted to the IEEE for possible publication
♻ ☆ GenSafe: A Generalizable Safety Enhancer for Safe Reinforcement Learning Algorithms Based on Reduced Order Markov Decision Process Model
Safe Reinforcement Learning (SRL) aims to realize a safe learning process for Deep Reinforcement Learning (DRL) algorithms by incorporating safety constraints. However, the efficacy of SRL approaches often relies on accurate function approximations, which are notably challenging to achieve in the early learning stages due to data insufficiency. To address this issue, we introduce in this work a novel Generalizable Safety enhancer (GenSafe) that is able to overcome the challenge of data insufficiency and enhance the performance of SRL approaches. Leveraging model order reduction techniques, we first propose an innovative method to construct a Reduced Order Markov Decision Process (ROMDP) as a low-dimensional approximator of the original safety constraints. Then, by solving the reformulated ROMDP-based constraints, GenSafe refines the actions of the agent to increase the possibility of constraint satisfaction. Essentially, GenSafe acts as an additional safety layer for SRL algorithms. We evaluate GenSafe on multiple SRL approaches and benchmark problems. The results demonstrate its capability to improve safety performance, especially in the early learning phases, while maintaining satisfactory task performance. Our proposed GenSafe not only offers a novel measure to augment existing SRL methods but also shows broad compatibility with various SRL algorithms, making it applicable to a wide range of systems and SRL problems.
♻ ☆ Perception Matters: Enhancing Embodied AI with Uncertainty-Aware Semantic Segmentation
Embodied AI has made significant progress acting in unexplored environments. However, tasks such as object search have largely focused on efficient policy learning. In this work, we identify several gaps in current search methods: They largely focus on dated perception models, neglect temporal aggregation, and transfer from ground truth directly to noisy perception at test time, without accounting for the resulting overconfidence in the perceived state. We address the identified problems through calibrated perception probabilities and uncertainty across aggregation and found decisions, thereby adapting the models for sequential tasks. The resulting methods can be directly integrated with pretrained models across a wide family of existing search approaches at no additional training cost. We perform extensive evaluations of aggregation methods across both different semantic perception models and policies, confirming the importance of calibrated uncertainties in both the aggregation and found decisions. We make the code and trained models available at https://semantic-search.cs.uni-freiburg.de.
♻ ☆ DIDLM: A SLAM Dataset for Difficult Scenarios Featuring Infrared, Depth Cameras, LIDAR, 4D Radar, and Others under Adverse Weather, Low Light Conditions, and Rough Roads
Adverse weather conditions, low-light environments, and bumpy road surfaces pose significant challenges to SLAM in robotic navigation and autonomous driving. Existing datasets in this field predominantly rely on single sensors or combinations of LiDAR, cameras, and IMUs. However, 4D millimeter-wave radar demonstrates robustness in adverse weather, infrared cameras excel in capturing details under low-light conditions, and depth images provide richer spatial information. Multi-sensor fusion methods also show potential for better adaptation to bumpy roads. Despite some SLAM studies incorporating these sensors and conditions, there remains a lack of comprehensive datasets addressing low-light environments and bumpy road conditions, or featuring a sufficiently diverse range of sensor data. In this study, we introduce a multi-sensor dataset covering challenging scenarios such as snowy weather, rainy weather, nighttime conditions, speed bumps, and rough terrains. The dataset includes rarely utilized sensors for extreme conditions, such as 4D millimeter-wave radar, infrared cameras, and depth cameras, alongside 3D LiDAR, RGB cameras, GPS, and IMU. It supports both autonomous driving and ground robot applications and provides reliable GPS/INS ground truth data, covering structured and semi-structured terrains. We evaluated various SLAM algorithms using this dataset, including RGB images, infrared images, depth images, LiDAR, and 4D millimeter-wave radar. The dataset spans a total of 18.5 km, 69 minutes, and approximately 660 GB, offering a valuable resource for advancing SLAM research under complex and extreme conditions. Our dataset is available at https://github.com/GongWeiSheng/DIDLM.
♻ ☆ Evaluation of Artificial Intelligence Methods for Lead Time Prediction in Non-Cycled Areas of Automotive Production
The present study examines the effectiveness of applying Artificial Intelligence methods in an automotive production environment to predict unknown lead times in a non-cycle-controlled production area. Data structures are analyzed to identify contextual features and then preprocessed using one-hot encoding. Methods selection focuses on supervised machine learning techniques. In supervised learning methods, regression and classification methods are evaluated. Continuous regression based on target size distribution is not feasible. Classification methods analysis shows that Ensemble Learning and Support Vector Machines are the most suitable. Preliminary study results indicate that gradient boosting algorithms LightGBM, XGBoost, and CatBoost yield the best results. After further testing and extensive hyperparameter optimization, the final method choice is the LightGBM algorithm. Depending on feature availability and prediction interval granularity, relative prediction accuracies of up to 90% can be achieved. Further tests highlight the importance of periodic retraining of AI models to accurately represent complex production processes using the database. The research demonstrates that AI methods can be effectively applied to highly variable production data, adding business value by providing an additional metric for various control tasks while outperforming current non AI-based systems.
♻ ☆ GazeGrasp: DNN-Driven Robotic Grasping with Wearable Eye-Gaze Interface
We present GazeGrasp, a gaze-based manipulation system enabling individuals with motor impairments to control collaborative robots using eye-gaze. The system employs an ESP32 CAM for eye tracking, MediaPipe for gaze detection, and YOLOv8 for object localization, integrated with a Universal Robot UR10 for manipulation tasks. After user-specific calibration, the system allows intuitive object selection with a magnetic snapping effect and robot control via eye gestures. Experimental evaluation involving 13 participants demonstrated that the magnetic snapping effect significantly reduced gaze alignment time, improving task efficiency by 31%. GazeGrasp provides a robust, hands-free interface for assistive robotics, enhancing accessibility and autonomy for users.
comment: Accepted to: IEEE/ACM International Conference on Human-Robot Interaction (HRI 2025)
♻ ☆ Cooperative Aerial Robot Inspection Challenge: A Benchmark for Heterogeneous Multi-UAV Planning and Lessons Learned
We propose the Cooperative Aerial Robot Inspection Challenge (CARIC), a simulation-based benchmark for motion planning algorithms in heterogeneous multi-UAV systems. CARIC features UAV teams with complementary sensors, realistic constraints, and evaluation metrics prioritizing inspection quality and efficiency. It offers a ready-to-use perception-control software stack and diverse scenarios to support the development and evaluation of task allocation and motion planning algorithms. Competitions using CARIC were held at IEEE CDC 2023 and the IROS 2024 Workshop on Multi-Robot Perception and Navigation, attracting innovative solutions from research teams worldwide. This paper examines the top three teams from CDC 2023, analyzing their exploration, inspection, and task allocation strategies while drawing insights into their performance across scenarios. The results highlight the task's complexity and suggest promising directions for future research in cooperative multi-UAV systems.
comment: Please find our website at https://ntu-aris.github.io/caric
♻ ☆ Analyzing Infrastructure LiDAR Placement with Realistic LiDAR Simulation Library ICRA'23
Recently, Vehicle-to-Everything(V2X) cooperative perception has attracted increasing attention. Infrastructure sensors play a critical role in this research field; however, how to find the optimal placement of infrastructure sensors is rarely studied. In this paper, we investigate the problem of infrastructure sensor placement and propose a pipeline that can efficiently and effectively find optimal installation positions for infrastructure sensors in a realistic simulated environment. To better simulate and evaluate LiDAR placement, we establish a Realistic LiDAR Simulation library that can simulate the unique characteristics of different popular LiDARs and produce high-fidelity LiDAR point clouds in the CARLA simulator. Through simulating point cloud data in different LiDAR placements, we can evaluate the perception accuracy of these placements using multiple detection models. Then, we analyze the correlation between the point cloud distribution and perception accuracy by calculating the density and uniformity of regions of interest. Experiments show that when using the same number and type of LiDAR, the placement scheme optimized by our proposed method improves the average precision by 15%, compared with the conventional placement scheme in the standard lane scene. We also analyze the correlation between perception performance in the region of interest and LiDAR point cloud distribution and validate that density and uniformity can be indicators of performance. Both the RLS Library and related code will be released at https://github.com/PJLab-ADG/PCSim.
comment: 7 pages, 6 figures, accepted to the IEEE International Conference on Robotics and Automation (ICRA'23)
♻ ☆ Cost-Effective Robotic Handwriting System with AI Integration
This paper introduces a cost-effective robotic handwriting system designed to replicate human-like handwriting with high precision. Combining a Raspberry Pi Pico microcontroller, 3D-printed components, and a machine learning-based handwriting generation model implemented via TensorFlow, the system converts user-supplied text into realistic stroke trajectories. By leveraging lightweight 3D-printed materials and efficient mechanical designs, the system achieves a total hardware cost of approximately \$56, significantly undercutting commercial alternatives. Experimental evaluations demonstrate handwriting precision within $\pm$0.3 millimeters and a writing speed of approximately 200 mm/min, positioning the system as a viable solution for educational, research, and assistive applications. This study seeks to lower the barriers to personalized handwriting technologies, making them accessible to a broader audience.
comment: This is an updated version of a paper originally presented at the 2024 IEEE Long Island Systems, Applications and Technology Conference (LISAT)
♻ ☆ Tactile-based Exploration, Mapping and Navigation with Collision-Resilient Aerial Vehicles
This article introduces XPLORER, a passive deformable UAV with a spring-augmented chassis and proprioceptive state awareness, designed to endure collisions and maintain smooth contact. We develop a fast-converging external force estimation algorithm for XPLORER that leverages onboard sensors and proprioceptive data for contact and collision detection. Using this force information, we propose four motion primitives, including three novel tactile-based primitives: tactile-traversal, tactile-turning, and ricocheting-to aid XPLORER in navigating unknown environments. These primitives are synthesized autonomously in real-time to enable efficient exploration and navigation by leveraging collisions and contacts. Experimental results demonstrate the effectiveness of our approach, highlighting the potential of passive deformable UAVs for contact-rich real-world tasks such as non-destructive inspection, surveillance and mapping, and pursuit/evasion.
♻ ☆ Safety Implications of Explainable Artificial Intelligence in End-to-End Autonomous Driving
The end-to-end learning pipeline is gradually creating a paradigm shift in the ongoing development of highly autonomous vehicles, largely due to advances in deep learning, the availability of large-scale training datasets, and improvements in integrated sensor devices. However, a lack of explainability in real-time decisions with contemporary learning methods impedes user trust and attenuates the widespread deployment and commercialization of such vehicles. Moreover, the issue is exacerbated when these cars are involved in or cause traffic accidents. Consequently, explainability in end-to-end autonomous driving is essential to build trust in vehicular automation. With that said, automotive researchers have not yet rigorously explored safety benefits and consequences of explanations in end-to-end autonomous driving. This paper aims to bridge the gaps between these topics and seeks to answer the following research question: What are safety implications of explanations in end-to-end autonomous driving? In this regard, we first revisit established safety and explainability concepts in end-to-end driving. Furthermore, we present three critical case studies and show the pivotal role of explanations in enhancing self-driving safety. Finally, we describe insights from empirical studies and reveal potential value, limitations, and caveats of practical explainable AI methods with respect to their safety assurance in end-to-end driving.
♻ ☆ Cooperative and Asynchronous Transformer-based Mission Planning for Heterogeneous Teams of Mobile Robots
Cooperative mission planning for heterogeneous teams of mobile robots presents a unique set of challenges, particularly when operating under communication constraints and limited computational resources. To address these challenges, we propose the Cooperative and Asynchronous Transformer-based Mission Planning (CATMiP) framework, which leverages multi-agent reinforcement learning (MARL) to coordinate distributed decision making among agents with diverse sensing, motion, and actuation capabilities, operating under sporadic ad hoc communication. A Class-based Macro-Action Decentralized Partially Observable Markov Decision Process (CMacDec-POMDP) is also formulated to effectively model asynchronous decision-making for heterogeneous teams of agents. The framework utilizes an asynchronous centralized training and distributed execution scheme that is developed based on the Multi-Agent Transformer (MAT) architecture. This design allows a single trained model to generalize to larger environments and accommodate varying team sizes and compositions. We evaluate CATMiP in a 2D grid-world simulation environment and compare its performance against planning-based exploration methods. Results demonstrate CATMiP's superior efficiency, scalability, and robustness to communication dropouts, highlighting its potential for real-world heterogeneous mobile robot systems. The code is available at https://github.com/mylad13/CATMiP.
comment: 27 pages, 8 figures, this work has been submitted to Elsevier for possible publication
♻ ☆ A Signal Temporal Logic Approach for Task-Based Coordination of Multi-Aerial Systems: a Wind Turbine Inspection Case Study
The paper addresses task assignment and trajectory generation for collaborative inspection missions using a fleet of multi-rotors, focusing on the wind turbine inspection scenario. The proposed solution enables safe and feasible trajectories while accommodating heterogeneous time-bound constraints and vehicle physical limits. An optimization problem is formulated to meet mission objectives and temporal requirements encoded as Signal Temporal Logic (STL) specifications. Additionally, an event-triggered replanner is introduced to address unforeseen events and compensate for lost time. Furthermore, a generalized robustness scoring method is employed to reflect user preferences and mitigate task conflicts. The effectiveness of the proposed approach is demonstrated through MATLAB and Gazebo simulations, as well as field multi-robot experiments in a mock-up scenario.
comment: \c{opyright}2025 Elsevier. This work has been accepted to "Robotics and Autonomous Systems" for possible publication. Personal use of this material is permitted. Permission from Elsevier must be obtained for all other uses
♻ ☆ Beyond Sight: Finetuning Generalist Robot Policies with Heterogeneous Sensors via Language Grounding
Interacting with the world is a multi-sensory experience: achieving effective general-purpose interaction requires making use of all available modalities -- including vision, touch, and audio -- to fill in gaps from partial observation. For example, when vision is occluded reaching into a bag, a robot should rely on its senses of touch and sound. However, state-of-the-art generalist robot policies are typically trained on large datasets to predict robot actions solely from visual and proprioceptive observations. In this work, we propose FuSe, a novel approach that enables finetuning visuomotor generalist policies on heterogeneous sensor modalities for which large datasets are not readily available by leveraging natural language as a common cross-modal grounding. We combine a multimodal contrastive loss with a sensory-grounded language generation loss to encode high-level semantics. In the context of robot manipulation, we show that FuSe enables performing challenging tasks that require reasoning jointly over modalities such as vision, touch, and sound in a zero-shot setting, such as multimodal prompting, compositional cross-modal prompting, and descriptions of objects it interacts with. We show that the same recipe is applicable to widely different generalist policies, including both diffusion-based generalist policies and large vision-language-action (VLA) models. Extensive experiments in the real world show that FuSeis able to increase success rates by over 20% compared to all considered baselines.
♻ ☆ Resilient Distributed Optimization for Multi-Agent Cyberphysical Systems
This work focuses on the problem of distributed optimization in multi-agent cyberphysical systems, where a legitimate agent's iterates are influenced both by the values it receives from potentially malicious neighboring agents, and by its own self-serving target function. We develop a new algorithmic and analytical framework to achieve resilience for the class of problems where stochastic values of trust between agents exist and can be exploited. In this case, we show that convergence to the true global optimal point can be recovered, both in mean and almost surely, even in the presence of malicious agents. Furthermore, we provide expected convergence rate guarantees in the form of upper bounds on the expected squared distance to the optimal value. Finally, numerical results are presented that validate our analytical convergence guarantees even when the malicious agents compose the majority of agents in the network and where existing methods fail to converge to the optimal nominal points.
comment: Accepted for publication in the IEEE Transactions on Automatic Control
♻ ☆ SYNAPSE: SYmbolic Neural-Aided Preference Synthesis Engine AAAI 25
This paper addresses the problem of preference learning, which aims to align robot behaviors through learning user specific preferences (e.g. "good pull-over location") from visual demonstrations. Despite its similarity to learning factual concepts (e.g. "red door"), preference learning is a fundamentally harder problem due to its subjective nature and the paucity of person-specific training data. We address this problem using a novel framework called SYNAPSE, which is a neuro-symbolic approach designed to efficiently learn preferential concepts from limited data. SYNAPSE represents preferences as neuro-symbolic programs, facilitating inspection of individual parts for alignment, in a domain-specific language (DSL) that operates over images and leverages a novel combination of visual parsing, large language models, and program synthesis to learn programs representing individual preferences. We perform extensive evaluations on various preferential concepts as well as user case studies demonstrating its ability to align well with dissimilar user preferences. Our method significantly outperforms baselines, especially when it comes to out of distribution generalization. We show the importance of the design choices in the framework through multiple ablation studies. Code, additional results, and supplementary material can be found on the website: https://amrl.cs.utexas.edu/synapse
comment: Accepted (oral) at AAAI 25
♻ ☆ GestLLM: Advanced Hand Gesture Interpretation via Large Language Models for Human-Robot Interaction
This paper introduces GestLLM, an advanced system for human-robot interaction that enables intuitive robot control through hand gestures. Unlike conventional systems, which rely on a limited set of predefined gestures, GestLLM leverages large language models and feature extraction via MediaPipe to interpret a diverse range of gestures. This integration addresses key limitations in existing systems, such as restricted gesture flexibility and the inability to recognize complex or unconventional gestures commonly used in human communication. By combining state-of-the-art feature extraction and language model capabilities, GestLLM achieves performance comparable to leading vision-language models while supporting gestures underrepresented in traditional datasets. For example, this includes gestures from popular culture, such as the ``Vulcan salute" from Star Trek, without any additional pretraining, prompt engineering, etc. This flexibility enhances the naturalness and inclusivity of robot control, making interactions more intuitive and user-friendly. GestLLM provides a significant step forward in gesture-based interaction, enabling robots to understand and respond to a wide variety of hand gestures effectively. This paper outlines its design, implementation, and evaluation, demonstrating its potential applications in advanced human-robot collaboration, assistive robotics, and interactive entertainment.
Multimedia 2
☆ Audio-visual Deepfake Detection With Local Temporal Inconsistencies ICASSP 2025
This paper proposes an audio-visual deepfake detection approach that aims to capture fine-grained temporal inconsistencies between audio and visual modalities. To achieve this, both architectural and data synthesis strategies are introduced. From an architectural perspective, a temporal distance map, coupled with an attention mechanism, is designed to capture these inconsistencies while minimizing the impact of irrelevant temporal subsequences. Moreover, we explore novel pseudo-fake generation techniques to synthesize local inconsistencies. Our approach is evaluated against state-of-the-art methods using the DFDC and FakeAVCeleb datasets, demonstrating its effectiveness in detecting audio-visual deepfakes.
comment: Accepted in ICASSP 2025
☆ Zero-shot Video Moment Retrieval via Off-the-shelf Multimodal Large Language Models AAAI 2025
The target of video moment retrieval (VMR) is predicting temporal spans within a video that semantically match a given linguistic query. Existing VMR methods based on multimodal large language models (MLLMs) overly rely on expensive high-quality datasets and time-consuming fine-tuning. Although some recent studies introduce a zero-shot setting to avoid fine-tuning, they overlook inherent language bias in the query, leading to erroneous localization. To tackle the aforementioned challenges, this paper proposes Moment-GPT, a tuning-free pipeline for zero-shot VMR utilizing frozen MLLMs. Specifically, we first employ LLaMA-3 to correct and rephrase the query to mitigate language bias. Subsequently, we design a span generator combined with MiniGPT-v2 to produce candidate spans adaptively. Finally, to leverage the video comprehension capabilities of MLLMs, we apply VideoChatGPT and span scorer to select the most appropriate spans. Our proposed method substantially outperforms the state-ofthe-art MLLM-based and zero-shot models on several public datasets, including QVHighlights, ActivityNet-Captions, and Charades-STA.
comment: Accepted by AAAI 2025
Computational Geometry 1
♻ ☆ Polycubes via Dual Loops
In this paper we study polycubes: orthogonal polyhedra with axis-aligned quadrilateral faces. We present a complete characterization of polycubes of any genus based on their dual structure: a collection of oriented loops which run in each of the axis directions and capture polycubes via their intersection patterns. A polycube loop structure uniquely corresponds to a polycube. We also describe all combinatorially different ways to add a loop to a loop structure while maintaining its validity. Similarly, we show how to identify loops that can be removed from a polycube loop structure without invalidating it. Our characterization gives rise to an iterative algorithm to construct provably valid polycube maps for a given input surface.
Computer Vision and Pattern Recognition 141
Dataset Distillation via Committee Voting
Dataset distillation aims to synthesize a smaller, representative dataset that preserves the essential properties of the original data, enabling efficient model training with reduced computational resources. Prior work has primarily focused on improving the alignment or matching process between original and synthetic data, or on enhancing the efficiency of distilling large datasets. In this work, we introduce ${\bf C}$ommittee ${\bf V}$oting for ${\bf D}$ataset ${\bf D}$istillation (CV-DD), a novel and orthogonal approach that leverages the collective wisdom of multiple models or experts to create high-quality distilled datasets. We start by showing how to establish a strong baseline that already achieves state-of-the-art accuracy through leveraging recent advancements and thoughtful adjustments in model design and optimization processes. By integrating distributions and predictions from a committee of models while generating high-quality soft labels, our method captures a wider spectrum of data features, reduces model-specific biases and the adverse effects of distribution shifts, leading to significant improvements in generalization. This voting-based strategy not only promotes diversity and robustness within the distilled dataset but also significantly reduces overfitting, resulting in improved performance on post-eval tasks. Extensive experiments across various datasets and IPCs (images per class) demonstrate that Committee Voting leads to more reliable and adaptable distilled data compared to single/multi-model distillation methods, demonstrating its potential for efficient and accurate dataset distillation. Code is available at: https://github.com/Jiacheng8/CV-DD.
comment: Code at: https://github.com/Jiacheng8/CV-DD
☆ UnCommon Objects in 3D
We introduce Uncommon Objects in 3D (uCO3D), a new object-centric dataset for 3D deep learning and 3D generative AI. uCO3D is the largest publicly-available collection of high-resolution videos of objects with 3D annotations that ensures full-360$^{\circ}$ coverage. uCO3D is significantly more diverse than MVImgNet and CO3Dv2, covering more than 1,000 object categories. It is also of higher quality, due to extensive quality checks of both the collected videos and the 3D annotations. Similar to analogous datasets, uCO3D contains annotations for 3D camera poses, depth maps and sparse point clouds. In addition, each object is equipped with a caption and a 3D Gaussian Splat reconstruction. We train several large 3D models on MVImgNet, CO3Dv2, and uCO3D and obtain superior results using the latter, showing that uCO3D is better for learning applications.
☆ Training-Free Motion-Guided Video Generation with Enhanced Temporal Consistency Using Motion Consistency Loss
In this paper, we address the challenge of generating temporally consistent videos with motion guidance. While many existing methods depend on additional control modules or inference-time fine-tuning, recent studies suggest that effective motion guidance is achievable without altering the model architecture or requiring extra training. Such approaches offer promising compatibility with various video generation foundation models. However, existing training-free methods often struggle to maintain consistent temporal coherence across frames or to follow guided motion accurately. In this work, we propose a simple yet effective solution that combines an initial-noise-based approach with a novel motion consistency loss, the latter being our key innovation. Specifically, we capture the inter-frame feature correlation patterns of intermediate features from a video diffusion model to represent the motion pattern of the reference video. We then design a motion consistency loss to maintain similar feature correlation patterns in the generated video, using the gradient of this loss in the latent space to guide the generation process for precise motion control. This approach improves temporal consistency across various motion control tasks while preserving the benefits of a training-free setup. Extensive experiments show that our method sets a new standard for efficient, temporally coherent video generation.
comment: Project page: https://zhangxinyu-xyz.github.io/SimulateMotion.github.io/
☆ MatchAnything: Universal Cross-Modality Image Matching with Large-Scale Pre-Training
Image matching, which aims to identify corresponding pixel locations between images, is crucial in a wide range of scientific disciplines, aiding in image registration, fusion, and analysis. In recent years, deep learning-based image matching algorithms have dramatically outperformed humans in rapidly and accurately finding large amounts of correspondences. However, when dealing with images captured under different imaging modalities that result in significant appearance changes, the performance of these algorithms often deteriorates due to the scarcity of annotated cross-modal training data. This limitation hinders applications in various fields that rely on multiple image modalities to obtain complementary information. To address this challenge, we propose a large-scale pre-training framework that utilizes synthetic cross-modal training signals, incorporating diverse data from various sources, to train models to recognize and match fundamental structures across images. This capability is transferable to real-world, unseen cross-modality image matching tasks. Our key finding is that the matching model trained with our framework achieves remarkable generalizability across more than eight unseen cross-modality registration tasks using the same network weight, substantially outperforming existing methods, whether designed for generalization or tailored for specific tasks. This advancement significantly enhances the applicability of image matching technologies across various scientific disciplines and paves the way for new applications in multi-modality human and artificial intelligence analysis and beyond.
comment: Project page: https://zju3dv.github.io/MatchAnything/
☆ SST-EM: Advanced Metrics for Evaluating Semantic, Spatial and Temporal Aspects in Video Editing WACV
Video editing models have advanced significantly, but evaluating their performance remains challenging. Traditional metrics, such as CLIP text and image scores, often fall short: text scores are limited by inadequate training data and hierarchical dependencies, while image scores fail to assess temporal consistency. We present SST-EM (Semantic, Spatial, and Temporal Evaluation Metric), a novel evaluation framework that leverages modern Vision-Language Models (VLMs), Object Detection, and Temporal Consistency checks. SST-EM comprises four components: (1) semantic extraction from frames using a VLM, (2) primary object tracking with Object Detection, (3) focused object refinement via an LLM agent, and (4) temporal consistency assessment using a Vision Transformer (ViT). These components are integrated into a unified metric with weights derived from human evaluations and regression analysis. The name SST-EM reflects its focus on Semantic, Spatial, and Temporal aspects of video evaluation. SST-EM provides a comprehensive evaluation of semantic fidelity and temporal smoothness in video editing. The source code is available in the \textbf{\href{https://github.com/custommetrics-sst/SST_CustomEvaluationMetrics.git}{GitHub Repository}}.
comment: WACV workshop
☆ Imagine while Reasoning in Space: Multimodal Visualization-of-Thought
Chain-of-Thought (CoT) prompting has proven highly effective for enhancing complex reasoning in Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs). Yet, it struggles in complex spatial reasoning tasks. Nonetheless, human cognition extends beyond language alone, enabling the remarkable capability to think in both words and images. Inspired by this mechanism, we propose a new reasoning paradigm, Multimodal Visualization-of-Thought (MVoT). It enables visual thinking in MLLMs by generating image visualizations of their reasoning traces. To ensure high-quality visualization, we introduce token discrepancy loss into autoregressive MLLMs. This innovation significantly improves both visual coherence and fidelity. We validate this approach through several dynamic spatial reasoning tasks. Experimental results reveal that MVoT demonstrates competitive performance across tasks. Moreover, it exhibits robust and reliable improvements in the most challenging scenarios where CoT fails. Ultimately, MVoT establishes new possibilities for complex reasoning tasks where visual thinking can effectively complement verbal reasoning.
comment: 11 pages, 6 figures, 4 tables (27 pages, 10 figures, 16 tables including references and appendices)
☆ Confident Pseudo-labeled Diffusion Augmentation for Canine Cardiomegaly Detection WACV
Canine cardiomegaly, marked by an enlarged heart, poses serious health risks if undetected, requiring accurate diagnostic methods. Current detection models often rely on small, poorly annotated datasets and struggle to generalize across diverse imaging conditions, limiting their real-world applicability. To address these issues, we propose a Confident Pseudo-labeled Diffusion Augmentation (CDA) model for identifying canine cardiomegaly. Our approach addresses the challenge of limited high-quality training data by employing diffusion models to generate synthetic X-ray images and annotate Vertebral Heart Score key points, thereby expanding the dataset. We also employ a pseudo-labeling strategy with Monte Carlo Dropout to select high-confidence labels, refine the synthetic dataset, and improve accuracy. Iteratively incorporating these labels enhances the model's performance, overcoming the limitations of existing approaches. Experimental results show that the CDA model outperforms traditional methods, achieving state-of-the-art accuracy in canine cardiomegaly detection. The code implementation is available at https://github.com/Shira7z/CDA.
comment: WACV workshop
☆ IP-FaceDiff: Identity-Preserving Facial Video Editing with Diffusion WACV-25
Facial video editing has become increasingly important for content creators, enabling the manipulation of facial expressions and attributes. However, existing models encounter challenges such as poor editing quality, high computational costs and difficulties in preserving facial identity across diverse edits. Additionally, these models are often constrained to editing predefined facial attributes, limiting their flexibility to diverse editing prompts. To address these challenges, we propose a novel facial video editing framework that leverages the rich latent space of pre-trained text-to-image (T2I) diffusion models and fine-tune them specifically for facial video editing tasks. Our approach introduces a targeted fine-tuning scheme that enables high quality, localized, text-driven edits while ensuring identity preservation across video frames. Additionally, by using pre-trained T2I models during inference, our approach significantly reduces editing time by 80%, while maintaining temporal consistency throughout the video sequence. We evaluate the effectiveness of our approach through extensive testing across a wide range of challenging scenarios, including varying head poses, complex action sequences, and diverse facial expressions. Our method consistently outperforms existing techniques, demonstrating superior performance across a broad set of metrics and benchmarks.
comment: WACV-25 Workshop
☆ RadAlign: Advancing Radiology Report Generation with Vision-Language Concept Alignment
Automated chest radiographs interpretation requires both accurate disease classification and detailed radiology report generation, presenting a significant challenge in the clinical workflow. Current approaches either focus on classification accuracy at the expense of interpretability or generate detailed but potentially unreliable reports through image captioning techniques. In this study, we present RadAlign, a novel framework that combines the predictive accuracy of vision-language models (VLMs) with the reasoning capabilities of large language models (LLMs). Inspired by the radiologist's workflow, RadAlign first employs a specialized VLM to align visual features with key medical concepts, achieving superior disease classification with an average AUC of 0.885 across multiple diseases. These recognized medical conditions, represented as text-based concepts in the aligned visual-language space, are then used to prompt LLM-based report generation. Enhanced by a retrieval-augmented generation mechanism that grounds outputs in similar historical cases, RadAlign delivers superior report quality with a GREEN score of 0.678, outperforming state-of-the-art methods' 0.634. Our framework maintains strong clinical interpretability while reducing hallucinations, advancing automated medical imaging and report analysis through integrated predictive and generative AI. Code is available at https://github.com/difeigu/RadAlign.
☆ Three-view Focal Length Recovery From Homographies
In this paper, we propose a novel approach for recovering focal lengths from three-view homographies. By examining the consistency of normal vectors between two homographies, we derive new explicit constraints between the focal lengths and homographies using an elimination technique. We demonstrate that three-view homographies provide two additional constraints, enabling the recovery of one or two focal lengths. We discuss four possible cases, including three cameras having an unknown equal focal length, three cameras having two different unknown focal lengths, three cameras where one focal length is known, and the other two cameras have equal or different unknown focal lengths. All the problems can be converted into solving polynomials in one or two unknowns, which can be efficiently solved using Sturm sequence or hidden variable technique. Evaluation using both synthetic and real data shows that the proposed solvers are both faster and more accurate than methods relying on existing two-view solvers. The code and data are available on https://github.com/kocurvik/hf
comment: Code available at https://github.com/kocurvik/hf Dataset available at: https://doi.org/10.5281/zenodo.14638904
☆ Aligning First, Then Fusing: A Novel Weakly Supervised Multimodal Violence Detection Method
Weakly supervised violence detection refers to the technique of training models to identify violent segments in videos using only video-level labels. Among these approaches, multimodal violence detection, which integrates modalities such as audio and optical flow, holds great potential. Existing methods in this domain primarily focus on designing multimodal fusion models to address modality discrepancies. In contrast, we take a different approach; leveraging the inherent discrepancies across modalities in violence event representation to propose a novel multimodal semantic feature alignment method. This method sparsely maps the semantic features of local, transient, and less informative modalities ( such as audio and optical flow ) into the more informative RGB semantic feature space. Through an iterative process, the method identifies the suitable no-zero feature matching subspace and aligns the modality-specific event representations based on this subspace, enabling the full exploitation of information from all modalities during the subsequent modality fusion stage. Building on this, we design a new weakly supervised violence detection framework that consists of unimodal multiple-instance learning for extracting unimodal semantic features, multimodal alignment, multimodal fusion, and final detection. Experimental results on benchmark datasets demonstrate the effectiveness of our method, achieving an average precision (AP) of 86.07% on the XD-Violence dataset. Our code is available at https://github.com/xjpp2016/MAVD.
☆ 3DGS-to-PC: Convert a 3D Gaussian Splatting Scene into a Dense Point Cloud or Mesh
3D Gaussian Splatting (3DGS) excels at producing highly detailed 3D reconstructions, but these scenes often require specialised renderers for effective visualisation. In contrast, point clouds are a widely used 3D representation and are compatible with most popular 3D processing software, yet converting 3DGS scenes into point clouds is a complex challenge. In this work we introduce 3DGS-to-PC, a flexible and highly customisable framework that is capable of transforming 3DGS scenes into dense, high-accuracy point clouds. We sample points probabilistically from each Gaussian as a 3D density function. We additionally threshold new points using the Mahalanobis distance to the Gaussian centre, preventing extreme outliers. The result is a point cloud that closely represents the shape encoded into the 3D Gaussian scene. Individual Gaussians use spherical harmonics to adapt colours depending on view, and each point may contribute only subtle colour hints to the resulting rendered scene. To avoid spurious or incorrect colours that do not fit with the final point cloud, we recalculate Gaussian colours via a customised image rendering approach, assigning each Gaussian the colour of the pixel to which it contributes most across all views. 3DGS-to-PC also supports mesh generation through Poisson Surface Reconstruction, applied to points sampled from predicted surface Gaussians. This allows coloured meshes to be generated from 3DGS scenes without the need for re-training. This package is highly customisable and capability of simple integration into existing 3DGS pipelines. 3DGS-to-PC provides a powerful tool for converting 3DGS data into point cloud and surface-based formats.
☆ A Survey on Dynamic Neural Networks: from Computer Vision to Multi-modal Sensor Fusion
Model compression is essential in the deployment of large Computer Vision models on embedded devices. However, static optimization techniques (e.g. pruning, quantization, etc.) neglect the fact that different inputs have different complexities, thus requiring different amount of computations. Dynamic Neural Networks allow to condition the number of computations to the specific input. The current literature on the topic is very extensive and fragmented. We present a comprehensive survey that synthesizes and unifies existing Dynamic Neural Networks research in the context of Computer Vision. Additionally, we provide a logical taxonomy based on which component of the network is adaptive: the output, the computation graph or the input. Furthermore, we argue that Dynamic Neural Networks are particularly beneficial in the context of Sensor Fusion for better adaptivity, noise reduction and information prioritization. We present preliminary works in this direction.
comment: Under review at International Journal of Computer Vision
☆ PrecipDiff: Leveraging image diffusion models to enhance satellite-based precipitation observations
A recent report from the World Meteorological Organization (WMO) highlights that water-related disasters have caused the highest human losses among natural disasters over the past 50 years, with over 91\% of deaths occurring in low-income countries. This disparity is largely due to the lack of adequate ground monitoring stations, such as weather surveillance radars (WSR), which are expensive to install. For example, while the US and Europe combined possess over 600 WSRs, Africa, despite having almost one and half times their landmass, has fewer than 40. To address this issue, satellite-based observations offer a global, near-real-time monitoring solution. However, they face several challenges like accuracy, bias, and low spatial resolution. This study leverages the power of diffusion models and residual learning to address these limitations in a unified framework. We introduce the first diffusion model for correcting the inconsistency between different precipitation products. Our method demonstrates the effectiveness in downscaling satellite precipitation estimates from 10 km to 1 km resolution. Extensive experiments conducted in the Seattle region demonstrate significant improvements in accuracy, bias reduction, and spatial detail. Importantly, our approach achieves these results using only precipitation data, showcasing the potential of a purely computer vision-based approach for enhancing satellite precipitation products and paving the way for further advancements in this domain.
☆ Guided SAM: Label-Efficient Part Segmentation
Localizing object parts precisely is essential for tasks such as object recognition and robotic manipulation. Recent part segmentation methods require extensive training data and labor-intensive annotations. Segment-Anything Model (SAM) has demonstrated good performance on a wide range of segmentation problems, but requires (manual) positional prompts to guide it where to segment. Furthermore, since it has been trained on full objects instead of object parts, it is prone to over-segmentation of parts. To address this, we propose a novel approach that guides SAM towards the relevant object parts. Our method learns positional prompts from coarse patch annotations that are easier and cheaper to acquire. We train classifiers on image patches to identify part classes and aggregate patches into regions of interest (ROIs) with positional prompts. SAM is conditioned on these ROIs and prompts. This approach, termed `Guided SAM', enhances efficiency and reduces manual effort, allowing effective part segmentation with minimal labeled data. We demonstrate the efficacy of Guided SAM on a dataset of car parts, improving the average IoU on state of the art models from 0.37 to 0.49 with annotations that are on average five times more efficient to acquire.
☆ Diff-Ensembler: Learning to Ensemble 2D Diffusion Models for Volume-to-Volume Medical Image Translation
Despite success in volume-to-volume translations in medical images, most existing models struggle to effectively capture the inherent volumetric distribution using 3D representations. The current state-of-the-art approach combines multiple 2D-based networks through weighted averaging, thereby neglecting the 3D spatial structures. Directly training 3D models in medical imaging presents significant challenges due to high computational demands and the need for large-scale datasets. To address these challenges, we introduce Diff-Ensembler, a novel hybrid 2D-3D model for efficient and effective volumetric translations by ensembling perpendicularly trained 2D diffusion models with a 3D network in each diffusion step. Moreover, our model can naturally be used to ensemble diffusion models conditioned on different modalities, allowing flexible and accurate fusion of input conditions. Extensive experiments demonstrate that Diff-Ensembler attains superior accuracy and volumetric realism in 3D medical image super-resolution and modality translation. We further demonstrate the strength of our model's volumetric realism using tumor segmentation as a downstream task.
☆ OCORD: Open-Campus Object Removal Dataset
The rapid advancements in generative models, particularly diffusion-based techniques, have revolutionized image inpainting tasks by enabling the generation of high-fidelity and diverse content. However, object removal remains under-explored as a specific subset of inpainting, facing challenges such as inadequate semantic understanding and the unintended generation of artifacts. Existing datasets for object removal often rely on synthetic data, which fails to align with real-world scenarios, limiting model performance. Although some real-world datasets address these issues partially, they suffer from scalability, annotation inefficiencies, and limited realism in physical phenomena such as lighting and shadows. To address these limitations, this paper introduces a novel approach to object removal by constructing a high-resolution real-world dataset through long-duration video capture with fixed camera settings. Leveraging advanced tools such as Grounding-DINO, Segment-Anything-Model, and MASA for automated annotation, we provides image, background, and mask pairs while significantly reducing annotation time and labor. With our efficient annotation pipeline, we release the first fully open, high-resolution real-world dataset for object removal, and improved performance in object removal tasks through fine-tuning of pre-trained diffusion models.
comment: technical report
☆ Zero-Shot Scene Understanding for Automatic Target Recognition Using Large Vision-Language Models
Automatic target recognition (ATR) plays a critical role in tasks such as navigation and surveillance, where safety and accuracy are paramount. In extreme use cases, such as military applications, these factors are often challenged due to the presence of unknown terrains, environmental conditions, and novel object categories. Current object detectors, including open-world detectors, lack the ability to confidently recognize novel objects or operate in unknown environments, as they have not been exposed to these new conditions. However, Large Vision-Language Models (LVLMs) exhibit emergent properties that enable them to recognize objects in varying conditions in a zero-shot manner. Despite this, LVLMs struggle to localize objects effectively within a scene. To address these limitations, we propose a novel pipeline that combines the detection capabilities of open-world detectors with the recognition confidence of LVLMs, creating a robust system for zero-shot ATR of novel classes and unknown domains. In this study, we compare the performance of various LVLMs for recognizing military vehicles, which are often underrepresented in training datasets. Additionally, we examine the impact of factors such as distance range, modality, and prompting methods on the recognition performance, providing insights into the development of more reliable ATR systems for novel conditions and classes.
☆ Kolmogorov-Arnold Network for Remote Sensing Image Semantic Segmentation
Semantic segmentation plays a crucial role in remote sensing applications, where the accurate extraction and representation of features are essential for high-quality results. Despite the widespread use of encoder-decoder architectures, existing methods often struggle with fully utilizing the high-dimensional features extracted by the encoder and efficiently recovering detailed information during decoding. To address these problems, we propose a novel semantic segmentation network, namely DeepKANSeg, including two key innovations based on the emerging Kolmogorov Arnold Network (KAN). Notably, the advantage of KAN lies in its ability to decompose high-dimensional complex functions into univariate transformations, enabling efficient and flexible representation of intricate relationships in data. First, we introduce a KAN-based deep feature refinement module, namely DeepKAN to effectively capture complex spatial and rich semantic relationships from high-dimensional features. Second, we replace the traditional multi-layer perceptron (MLP) layers in the global-local combined decoder with KAN-based linear layers, namely GLKAN. This module enhances the decoder's ability to capture fine-grained details during decoding. To evaluate the effectiveness of the proposed method, experiments are conducted on two well-known fine-resolution remote sensing benchmark datasets, namely ISPRS Vaihingen and ISPRS Potsdam. The results demonstrate that the KAN-enhanced segmentation model achieves superior performance in terms of accuracy compared to state-of-the-art methods. They highlight the potential of KANs as a powerful alternative to traditional architectures in semantic segmentation tasks. Moreover, the explicit univariate decomposition provides improved interpretability, which is particularly beneficial for applications requiring explainable learning in remote sensing.
comment: 13 pages, 8 figures
☆ FedSemiDG: Domain Generalized Federated Semi-supervised Medical Image Segmentation
Medical image segmentation is challenging due to the diversity of medical images and the lack of labeled data, which motivates recent developments in federated semi-supervised learning (FSSL) to leverage a large amount of unlabeled data from multiple centers for model training without sharing raw data. However, what remains under-explored in FSSL is the domain shift problem which may cause suboptimal model aggregation and low effectivity of the utilization of unlabeled data, eventually leading to unsatisfactory performance in unseen domains. In this paper, we explore this previously ignored scenario, namely domain generalized federated semi-supervised learning (FedSemiDG), which aims to learn a model in a distributed manner from multiple domains with limited labeled data and abundant unlabeled data such that the model can generalize well to unseen domains. We present a novel framework, Federated Generalization-Aware SemiSupervised Learning (FGASL), to address the challenges in FedSemiDG by effectively tackling critical issues at both global and local levels. Globally, we introduce Generalization-Aware Aggregation (GAA), assigning adaptive weights to local models based on their generalization performance. Locally, we use a Dual-Teacher Adaptive Pseudo Label Refinement (DR) strategy to combine global and domain-specific knowledge, generating more reliable pseudo labels. Additionally, Perturbation-Invariant Alignment (PIA) enforces feature consistency under perturbations, promoting domain-invariant learning. Extensive experiments on three medical segmentation tasks (cardiac MRI, spine MRI and bladder cancer MRI) demonstrate that our method significantly outperforms state-of-the-art FSSL and domain generalization approaches, achieving robust generalization on unseen domains.
comment: 17 pages
☆ TimberVision: A Multi-Task Dataset and Framework for Log-Component Segmentation and Tracking in Autonomous Forestry Operations WACV
Timber represents an increasingly valuable and versatile resource. However, forestry operations such as harvesting, handling and measuring logs still require substantial human labor in remote environments posing significant safety risks. Progressively automating these tasks has the potential of increasing their efficiency as well as safety, but requires an accurate detection of individual logs as well as live trees and their context. Although initial approaches have been proposed for this challenging application domain, specialized data and algorithms are still too scarce to develop robust solutions. To mitigate this gap, we introduce the TimberVision dataset, consisting of more than 2k annotated RGB images containing a total of 51k trunk components including cut and lateral surfaces, thereby surpassing any existing dataset in this domain in terms of both quantity and detail by a large margin. Based on this data, we conduct a series of ablation experiments for oriented object detection and instance segmentation and evaluate the influence of multiple scene parameters on model performance. We introduce a generic framework to fuse the components detected by our models for both tasks into unified trunk representations. Furthermore, we automatically derive geometric properties and apply multi-object tracking to further enhance robustness. Our detection and tracking approach provides highly descriptive and accurate trunk representations solely from RGB image data, even under challenging environmental conditions. Our solution is suitable for a wide range of application scenarios and can be readily combined with other sensor modalities.
comment: Accepted at Winter Conference on Applications of Computer Vision (WACV) 2025. Code and dataset available at https://github.com/timbervision/timbervision
☆ A method for estimating roadway billboard salience
Roadside billboards and other forms of outdoor advertising play a crucial role in marketing initiatives; however, they can also distract drivers, potentially contributing to accidents. This study delves into the significance of roadside advertising in images captured from a driver's perspective. Firstly, it evaluates the effectiveness of neural networks in detecting advertising along roads, focusing on the YOLOv5 and Faster R-CNN models. Secondly, the study addresses the determination of billboard significance using methods for saliency extraction. The UniSal and SpectralResidual methods were employed to create saliency maps for each image. The study establishes a database of eye tracking sessions captured during city highway driving to assess the saliency models.
☆ Anonymization of Documents for Law Enforcement with Machine Learning
The steadily increasing utilization of data-driven methods and approaches in areas that handle sensitive personal information such as in law enforcement mandates an ever increasing effort in these institutions to comply with data protection guidelines. In this work, we present a system for automatically anonymizing images of scanned documents, reducing manual effort while ensuring data protection compliance. Our method considers the viability of further forensic processing after anonymization by minimizing automatically redacted areas by combining automatic detection of sensitive regions with knowledge from a manually anonymized reference document. Using a self-supervised image model for instance retrieval of the reference document, our approach requires only one anonymized example to efficiently redact all documents of the same type, significantly reducing processing time. We show that our approach outperforms both a purely automatic redaction system and also a naive copy-paste scheme of the reference anonymization to other documents on a hand-crafted dataset of ground truth redactions.
comment: Accepted at IEEE Symposium on CI in Security, Defence and Biometrics 2025 (IEEE CISDB)
☆ Localization-Aware Multi-Scale Representation Learning for Repetitive Action Counting
Repetitive action counting (RAC) aims to estimate the number of class-agnostic action occurrences in a video without exemplars. Most current RAC methods rely on a raw frame-to-frame similarity representation for period prediction. However, this approach can be significantly disrupted by common noise such as action interruptions and inconsistencies, leading to sub-optimal counting performance in realistic scenarios. In this paper, we introduce a foreground localization optimization objective into similarity representation learning to obtain more robust and efficient video features. We propose a Localization-Aware Multi-Scale Representation Learning (LMRL) framework. Specifically, we apply a Multi-Scale Period-Aware Representation (MPR) with a scale-specific design to accommodate various action frequencies and learn more flexible temporal correlations. Furthermore, we introduce the Repetition Foreground Localization (RFL) method, which enhances the representation by coarsely identifying periodic actions and incorporating global semantic information. These two modules can be jointly optimized, resulting in a more discerning periodic action representation. Our approach significantly reduces the impact of noise, thereby improving counting accuracy. Additionally, the framework is designed to be scalable and adaptable to different types of video content. Experimental results on the RepCountA and UCFRep datasets demonstrate that our proposed method effectively handles repetitive action counting.
comment: Accepted by IEEE VCIP2024
☆ The Devil is in the Spurious Correlation: Boosting Moment Retrieval via Temporal Dynamic Learning
Given a textual query along with a corresponding video, the objective of moment retrieval aims to localize the moments relevant to the query within the video. While commendable results have been demonstrated by existing transformer-based approaches, predicting the accurate temporal span of the target moment is currently still a major challenge. In this paper, we reveal that a crucial reason stems from the spurious correlation between the text queries and the moment context. Namely, the model may associate the textual query with the background frames rather than the target moment. To address this issue, we propose a temporal dynamic learning approach for moment retrieval, where two strategies are designed to mitigate the spurious correlation. First, we introduce a novel video synthesis approach to construct a dynamic context for the relevant moment. With separate yet similar videos mixed up, the synthesis approach empowers our model to attend to the target moment of the corresponding query under various dynamic contexts. Second, we enhance the representation by learning temporal dynamics. Besides the visual representation, text queries are aligned with temporal dynamic representations, which enables our model to establish a non-spurious correlation between the query-related moment and context. With the aforementioned proposed method, the spurious correlation issue in moment retrieval can be largely alleviated. Our method establishes a new state-of-the-art performance on two popular benchmarks of moment retrieval, \ie, QVHighlights and Charades-STA. In addition, the detailed ablation analyses demonstrate the effectiveness of the proposed strategies. Our code will be publicly available.
☆ Code and Pixels: Multi-Modal Contrastive Pre-training for Enhanced Tabular Data Analysis
Learning from tabular data is of paramount importance, as it complements the conventional analysis of image and video data by providing a rich source of structured information that is often critical for comprehensive understanding and decision-making processes. We present Multi-task Contrastive Masked Tabular Modeling (MT-CMTM), a novel method aiming to enhance tabular models by leveraging the correlation between tabular data and corresponding images. MT-CMTM employs a dual strategy combining contrastive learning with masked tabular modeling, optimizing the synergy between these data modalities. Central to our approach is a 1D Convolutional Neural Network with residual connections and an attention mechanism (1D-ResNet-CBAM), designed to efficiently process tabular data without relying on images. This enables MT-CMTM to handle purely tabular data for downstream tasks, eliminating the need for potentially costly image acquisition and processing. We evaluated MT-CMTM on the DVM car dataset, which is uniquely suited for this particular scenario, and the newly developed HIPMP dataset, which connects membrane fabrication parameters with image data. Our MT-CMTM model outperforms the proposed tabular 1D-ResNet-CBAM, which is trained from scratch, achieving a relative 1.48% improvement in relative MSE on HIPMP and a 2.38% increase in absolute accuracy on DVM. These results demonstrate MT-CMTM's robustness and its potential to advance the field of multi-modal learning.
☆ Comparative analysis of optical character recognition methods for Sámi texts from the National Library of Norway
Optical Character Recognition (OCR) is crucial to the National Library of Norway's (NLN) digitisation process as it converts scanned documents into machine-readable text. However, for the S\'ami documents in NLN's collection, the OCR accuracy is insufficient. Given that OCR quality affects downstream processes, evaluating and improving OCR for text written in S\'ami languages is necessary to make these resources accessible. To address this need, this work fine-tunes and evaluates three established OCR approaches, Transkribus, Tesseract and TrOCR, for transcribing S\'ami texts from NLN's collection. Our results show that Transkribus and TrOCR outperform Tesseract on this task, while Tesseract achieves superior performance on an out-of-domain dataset. Furthermore, we show that fine-tuning pre-trained models and supplementing manual annotations with machine annotations and synthetic text images can yield accurate OCR for S\'ami languages, even with a moderate amount of manually annotated data.
comment: To be published in Proceedings of the 25th Nordic Conference on Computational Linguistics (NoDaLiDa)
☆ Toward Realistic Camouflaged Object Detection: Benchmarks and Method
Camouflaged object detection (COD) primarily relies on semantic or instance segmentation methods. While these methods have made significant advancements in identifying the contours of camouflaged objects, they may be inefficient or cost-effective for tasks that only require the specific location of the object. Object detection algorithms offer an optimized solution for Realistic Camouflaged Object Detection (RCOD) in such cases. However, detecting camouflaged objects remains a formidable challenge due to the high degree of similarity between the features of the objects and their backgrounds. Unlike segmentation methods that perform pixel-wise comparisons to differentiate between foreground and background, object detectors omit this analysis, further aggravating the challenge. To solve this problem, we propose a camouflage-aware feature refinement (CAFR) strategy. Since camouflaged objects are not rare categories, CAFR fully utilizes a clear perception of the current object within the prior knowledge of large models to assist detectors in deeply understanding the distinctions between background and foreground. Specifically, in CAFR, we introduce the Adaptive Gradient Propagation (AGP) module that fine-tunes all feature extractor layers in large detection models to fully refine class-specific features from camouflaged contexts. We then design the Sparse Feature Refinement (SFR) module that optimizes the transformer-based feature extractor to focus primarily on capturing class-specific features in camouflaged scenarios. To facilitate the assessment of RCOD tasks, we manually annotate the labels required for detection on three existing segmentation COD datasets, creating a new benchmark for RCOD tasks. Code and datasets are available at: https://github.com/zhimengXin/RCOD.
☆ Event-based Video Person Re-identification via Cross-Modality and Temporal Collaboration ICASSP 2025
Video-based person re-identification (ReID) has become increasingly important due to its applications in video surveillance applications. By employing events in video-based person ReID, more motion information can be provided between continuous frames to improve recognition accuracy. Previous approaches have assisted by introducing event data into the video person ReID task, but they still cannot avoid the privacy leakage problem caused by RGB images. In order to avoid privacy attacks and to take advantage of the benefits of event data, we consider using only event data. To make full use of the information in the event stream, we propose a Cross-Modality and Temporal Collaboration (CMTC) network for event-based video person ReID. First, we design an event transform network to obtain corresponding auxiliary information from the input of raw events. Additionally, we propose a differential modality collaboration module to balance the roles of events and auxiliaries to achieve complementary effects. Furthermore, we introduce a temporal collaboration module to exploit motion information and appearance cues. Experimental results demonstrate that our method outperforms others in the task of event-based video person ReID.
comment: Accepted by ICASSP 2025
☆ Skip Mamba Diffusion for Monocular 3D Semantic Scene Completion AAAI 2025
3D semantic scene completion is critical for multiple downstream tasks in autonomous systems. It estimates missing geometric and semantic information in the acquired scene data. Due to the challenging real-world conditions, this task usually demands complex models that process multi-modal data to achieve acceptable performance. We propose a unique neural model, leveraging advances from the state space and diffusion generative modeling to achieve remarkable 3D semantic scene completion performance with monocular image input. Our technique processes the data in the conditioned latent space of a variational autoencoder where diffusion modeling is carried out with an innovative state space technique. A key component of our neural network is the proposed Skimba (Skip Mamba) denoiser, which is adept at efficiently processing long-sequence data. The Skimba diffusion model is integral to our 3D scene completion network, incorporating a triple Mamba structure, dimensional decomposition residuals and varying dilations along three directions. We also adopt a variant of this network for the subsequent semantic segmentation stage of our method. Extensive evaluation on the standard SemanticKITTI and SSCBench-KITTI360 datasets show that our approach not only outperforms other monocular techniques by a large margin, it also achieves competitive performance against stereo methods. The code is available at https://github.com/xrkong/skimba
comment: Accepted by AAAI 2025
☆ EdgeTAM: On-Device Track Anything Model
On top of Segment Anything Model (SAM), SAM 2 further extends its capability from image to video inputs through a memory bank mechanism and obtains a remarkable performance compared with previous methods, making it a foundation model for video segmentation task. In this paper, we aim at making SAM 2 much more efficient so that it even runs on mobile devices while maintaining a comparable performance. Despite several works optimizing SAM for better efficiency, we find they are not sufficient for SAM 2 because they all focus on compressing the image encoder, while our benchmark shows that the newly introduced memory attention blocks are also the latency bottleneck. Given this observation, we propose EdgeTAM, which leverages a novel 2D Spatial Perceiver to reduce the computational cost. In particular, the proposed 2D Spatial Perceiver encodes the densely stored frame-level memories with a lightweight Transformer that contains a fixed set of learnable queries. Given that video segmentation is a dense prediction task, we find preserving the spatial structure of the memories is essential so that the queries are split into global-level and patch-level groups. We also propose a distillation pipeline that further improves the performance without inference overhead. As a result, EdgeTAM achieves 87.7, 70.0, 72.3, and 71.7 J&F on DAVIS 2017, MOSE, SA-V val, and SA-V test, while running at 16 FPS on iPhone 15 Pro Max.
comment: Code will be released at https://github.com/facebookresearch/EdgeTAM
☆ MOS-Attack: A Scalable Multi-objective Adversarial Attack Framework CVPR 2025
Crafting adversarial examples is crucial for evaluating and enhancing the robustness of Deep Neural Networks (DNNs), presenting a challenge equivalent to maximizing a non-differentiable 0-1 loss function. However, existing single objective methods, namely adversarial attacks focus on a surrogate loss function, do not fully harness the benefits of engaging multiple loss functions, as a result of insufficient understanding of their synergistic and conflicting nature. To overcome these limitations, we propose the Multi-Objective Set-based Attack (MOS Attack), a novel adversarial attack framework leveraging multiple loss functions and automatically uncovering their interrelations. The MOS Attack adopts a set-based multi-objective optimization strategy, enabling the incorporation of numerous loss functions without additional parameters. It also automatically mines synergistic patterns among various losses, facilitating the generation of potent adversarial attacks with fewer objectives. Extensive experiments have shown that our MOS Attack outperforms single-objective attacks. Furthermore, by harnessing the identified synergistic patterns, MOS Attack continues to show superior results with a reduced number of loss functions.
comment: Under Review of CVPR 2025
☆ Implicit Neural Representations for Registration of Left Ventricle Myocardium During a Cardiac Cycle
Understanding the movement of the left ventricle myocardium (LVmyo) during the cardiac cycle is essential for assessing cardiac function. One way to model this movement is through a series of deformable image registrations (DIRs) of the LVmyo. Traditional deep learning methods for DIRs, such as those based on convolutional neural networks, often require substantial memory and computational resources. In contrast, implicit neural representations (INRs) offer an efficient approach by operating on any number of continuous points. This study extends the use of INRs for DIR to cardiac computed tomography (CT), focusing on LVmyo registration. To enhance the precision of the registration around the LVmyo, we incorporate the signed distance field of the LVmyo with the Hounsfield Unit values from the CT frames. This guides the registration of the LVmyo, while keeping the tissue information from the CT frames. Our framework demonstrates high registration accuracy and provides a robust method for temporal registration that facilitates further analysis of LVmyo motion.
comment: 9 pages, 5 figures, STACOM 2024
☆ Depth and Image Fusion for Road Obstacle Detection Using Stereo Camera
This paper is devoted to the detection of objects on a road, performed with a combination of two methods based on both the use of depth information and video analysis of data from a stereo camera. Since neither the time of the appearance of an object on the road, nor its size and shape is known in advance, ML/DL-based approaches are not applicable. The task becomes more complicated due to variations in artificial illumination, inhomogeneous road surface texture, and unknown character and features of the object. To solve this problem we developed the depth and image fusion method that complements a search of small contrast objects by RGB-based method, and obstacle detection by stereo image-based approach with SLIC superpixel segmentation. We conducted experiments with static and low speed obstacles in an underground parking lot and demonstrated the successful work of the developed technique for detecting and even tracking small objects, which can be parking infrastructure objects, things left on the road, wheels, dropped boxes, etc.
comment: 8 pages, 15 figures
☆ Can Vision-Language Models Evaluate Handwritten Math?
Recent advancements in Vision-Language Models (VLMs) have opened new possibilities in automatic grading of handwritten student responses, particularly in mathematics. However, a comprehensive study to test the ability of VLMs to evaluate and reason over handwritten content remains absent. To address this gap, we introduce FERMAT, a benchmark designed to assess the ability of VLMs to detect, localize and correct errors in handwritten mathematical content. FERMAT spans four key error dimensions - computational, conceptual, notational, and presentation - and comprises over 2,200 handwritten math solutions derived from 609 manually curated problems from grades 7-12 with intentionally introduced perturbations. Using FERMAT we benchmark nine VLMs across three tasks: error detection, localization, and correction. Our results reveal significant shortcomings in current VLMs in reasoning over handwritten text, with Gemini-1.5-Pro achieving the highest error correction rate (77%). We also observed that some models struggle with processing handwritten content, as their accuracy improves when handwritten inputs are replaced with printed text or images. These findings highlight the limitations of current VLMs and reveal new avenues for improvement. We release FERMAT and all the associated resources in the open-source to drive further research.
☆ CSTA: Spatial-Temporal Causal Adaptive Learning for Exemplar-Free Video Class-Incremental Learning
Continual learning aims to acquire new knowledge while retaining past information. Class-incremental learning (CIL) presents a challenging scenario where classes are introduced sequentially. For video data, the task becomes more complex than image data because it requires learning and preserving both spatial appearance and temporal action involvement. To address this challenge, we propose a novel exemplar-free framework that equips separate spatiotemporal adapters to learn new class patterns, accommodating the incremental information representation requirements unique to each class. While separate adapters are proven to mitigate forgetting and fit unique requirements, naively applying them hinders the intrinsic connection between spatial and temporal information increments, affecting the efficiency of representing newly learned class information. Motivated by this, we introduce two key innovations from a causal perspective. First, a causal distillation module is devised to maintain the relation between spatial-temporal knowledge for a more efficient representation. Second, a causal compensation mechanism is proposed to reduce the conflicts during increment and memorization between different types of information. Extensive experiments conducted on benchmark datasets demonstrate that our framework can achieve new state-of-the-art results, surpassing current example-based methods by 4.2% in accuracy on average.
comment: IEEE TCSVT Submission
☆ MECD+: Unlocking Event-Level Causal Graph Discovery for Video Reasoning
Video causal reasoning aims to achieve a high-level understanding of videos from a causal perspective. However, it exhibits limitations in its scope, primarily executed in a question-answering paradigm and focusing on brief video segments containing isolated events and basic causal relations, lacking comprehensive and structured causality analysis for videos with multiple interconnected events. To fill this gap, we introduce a new task and dataset, Multi-Event Causal Discovery (MECD). It aims to uncover the causal relations between events distributed chronologically across long videos. Given visual segments and textual descriptions of events, MECD identifies the causal associations between these events to derive a comprehensive and structured event-level video causal graph explaining why and how the result event occurred. To address the challenges of MECD, we devise a novel framework inspired by the Granger Causality method, incorporating an efficient mask-based event prediction model to perform an Event Granger Test. It estimates causality by comparing the predicted result event when premise events are masked versus unmasked. Furthermore, we integrate causal inference techniques such as front-door adjustment and counterfactual inference to mitigate challenges in MECD like causality confounding and illusory causality. Additionally, context chain reasoning is introduced to conduct more robust and generalized reasoning. Experiments validate the effectiveness of our framework in reasoning complete causal relations, outperforming GPT-4o and VideoChat2 by 5.77% and 2.70%, respectively. Further experiments demonstrate that causal relation graphs can also contribute to downstream video understanding tasks such as video question answering and video event prediction.
comment: IEEE TPAMI Submission. arXiv admin note: substantial text overlap with arXiv:2409.17647
☆ Exploring the Use of Contrastive Language-Image Pre-Training for Human Posture Classification: Insights from Yoga Pose Analysis
Accurate human posture classification in images and videos is crucial for automated applications across various fields, including work safety, physical rehabilitation, sports training, or daily assisted living. Recently, multimodal learning methods, such as Contrastive Language-Image Pretraining (CLIP), have advanced significantly in jointly understanding images and text. This study aims to assess the effectiveness of CLIP in classifying human postures, focusing on its application in yoga. Despite the initial limitations of the zero-shot approach, applying transfer learning on 15,301 images (real and synthetic) with 82 classes has shown promising results. The article describes the full procedure for fine-tuning, including the choice for image description syntax, models and hyperparameters adjustment. The fine-tuned CLIP model, tested on 3826 images, achieves an accuracy of over 85%, surpassing the current state-of-the-art of previous works on the same dataset by approximately 6%, its training time being 3.5 times lower than what is needed to fine-tune a YOLOv8-based model. For more application-oriented scenarios, with smaller datasets of six postures each, containing 1301 and 401 training images, the fine-tuned models attain an accuracy of 98.8% and 99.1%, respectively. Furthermore, our experiments indicate that training with as few as 20 images per pose can yield around 90% accuracy in a six-class dataset. This study demonstrates that this multimodal technique can be effectively used for yoga pose classification, and possibly for human posture classification, in general. Additionally, CLIP inference time (around 7 ms) supports that the model can be integrated into automated systems for posture evaluation, e.g., for developing a real-time personal yoga assistant for performance assessment.
☆ TimeLogic: A Temporal Logic Benchmark for Video QA
Temporal logical understanding, a core facet of human cognition, plays a pivotal role in capturing complex sequential events and their temporal relationships within videos. This capability is particularly crucial in tasks like Video Question Answering (VideoQA), where the goal is to process visual data over time together with textual data to provide coherent answers. However, current VideoQA benchmarks devote little focus to evaluating this critical skill due to the challenge of annotating temporal logic. Despite the advancement of vision-language models, assessing their temporal logical reasoning powers remains a challenge, primarily due to the lack QA pairs that demand formal, complex temporal reasoning. To bridge this gap, we introduce the TimeLogic QA (TLQA) framework to automatically generate the QA pairs, specifically designed to evaluate the temporal logical understanding. To this end, TLQA leverages temporal annotations from existing video datasets together with temporal operators derived from logic theory to construct questions that test understanding of event sequences and their temporal relationships. TLQA framework is generic and scalable, capable of leveraging both, existing video action datasets with temporal action segmentation annotations, or video datasets with temporal scene graph annotations, to automatically generate temporal logical questions. We leverage 4 datasets, STAR, Breakfast, AGQA, and CrossTask, and generate two VideoQA dataset variants - small (TLQA-S) and large (TLQA-L) - containing 2k and 10k QA pairs for each category, resulting in 32k and 160k total pairs per dataset. We undertake a comprehensive evaluation of leading-edge VideoQA models, employing the TLQA to benchmark their temporal logical understanding capabilities. We assess the VideoQA model's temporal reasoning performance on 16 categories of temporal logic with varying temporal complexity.
☆ Multi-face emotion detection for effective Human-Robot Interaction
The integration of dialogue interfaces in mobile devices has become ubiquitous, providing a wide array of services. As technology progresses, humanoid robots designed with human-like features to interact effectively with people are gaining prominence, and the use of advanced human-robot dialogue interfaces is continually expanding. In this context, emotion recognition plays a crucial role in enhancing human-robot interaction by enabling robots to understand human intentions. This research proposes a facial emotion detection interface integrated into a mobile humanoid robot, capable of displaying real-time emotions from multiple individuals on a user interface. To this end, various deep neural network models for facial expression recognition were developed and evaluated under consistent computer-based conditions, yielding promising results. Afterwards, a trade-off between accuracy and memory footprint was carefully considered to effectively implement this application on a mobile humanoid robot.
comment: 9 pages, 8 figures and 1 table. Accepted at the 17th International Conference on Agents and Artificial Intelligence (ICAART 2025), Porto, Portugal
☆ FaceOracle: Chat with a Face Image Oracle
A face image is a mandatory part of ID and travel documents. Obtaining high-quality face images when issuing such documents is crucial for both human examiners and automated face recognition systems. In several international standards, face image quality requirements are intricate and defined in detail. Identifying and understanding non-compliance or defects in the submitted face images is crucial for both issuing authorities and applicants. In this work, we introduce FaceOracle, an LLM-powered AI assistant that helps its users analyze a face image in a natural conversational manner using standard compliant algorithms. Leveraging the power of LLMs, users can get explanations of various face image quality concepts as well as interpret the outcome of face image quality assessment (FIQA) algorithms. We implement a proof-of-concept that demonstrates how experts at an issuing authority could integrate FaceOracle into their workflow to analyze, understand, and communicate their decisions more efficiently, resulting in enhanced productivity.
☆ Lung Cancer detection using Deep Learning
In this paper we discuss lung cancer detection using hybrid model of Convolutional-Neural-Networks (CNNs) and Support-Vector-Machines-(SVMs) in order to gain early detection of tumors, benign or malignant. The work uses this hybrid model by training upon the Computed Tomography scans (CT scans) as dataset. Using deep learning for detecting lung cancer early is a cutting-edge method.
☆ VAGeo: View-specific Attention for Cross-View Object Geo-Localization ICASSP 2025
Cross-view object geo-localization (CVOGL) aims to locate an object of interest in a captured ground- or drone-view image within the satellite image. However, existing works treat ground-view and drone-view query images equivalently, overlooking their inherent viewpoint discrepancies and the spatial correlation between the query image and the satellite-view reference image. To this end, this paper proposes a novel View-specific Attention Geo-localization method (VAGeo) for accurate CVOGL. Specifically, VAGeo contains two key modules: view-specific positional encoding (VSPE) module and channel-spatial hybrid attention (CSHA) module. In object-level, according to the characteristics of different viewpoints of ground and drone query images, viewpoint-specific positional codings are designed to more accurately identify the click-point object of the query image in the VSPE module. In feature-level, a hybrid attention in the CSHA module is introduced by combining channel attention and spatial attention mechanisms simultaneously for learning discriminative features. Extensive experimental results demonstrate that the proposed VAGeo gains a significant performance improvement, i.e., improving acc@0.25/acc@0.5 on the CVOGL dataset from 45.43%/42.24% to 48.21%/45.22% for ground-view, and from 61.97%/57.66% to 66.19%/61.87% for drone-view.
comment: Accepted by ICASSP 2025
☆ A4O: All Trigger for One sample
Backdoor attacks have become a critical threat to deep neural networks (DNNs), drawing many research interests. However, most of the studied attacks employ a single type of trigger. Consequently, proposed backdoor defenders often rely on the assumption that triggers would appear in a unified way. In this paper, we show that this naive assumption can create a loophole, allowing more sophisticated backdoor attacks to bypass. We design a novel backdoor attack mechanism that incorporates multiple types of backdoor triggers, focusing on stealthiness and effectiveness. Our journey begins with the intriguing observation that the performance of a backdoor attack in deep learning models, as well as its detectability and removability, are all proportional to the magnitude of the trigger. Based on this correlation, we propose reducing the magnitude of each trigger type and combining them to achieve a strong backdoor relying on the combined trigger while still staying safely under the radar of defenders. Extensive experiments on three standard datasets demonstrate that our method can achieve high attack success rates (ASRs) while consistently bypassing state-of-the-art defenses.
☆ Uncertainty Guarantees on Automated Precision Weeding using Conformal Prediction
Precision agriculture in general, and precision weeding in particular, have greatly benefited from the major advancements in deep learning and computer vision. A large variety of commercial robotic solutions are already available and deployed. However, the adoption by farmers of such solutions is still low for many reasons, an important one being the lack of trust in these systems. This is in great part due to the opaqueness and complexity of deep neural networks and the manufacturers' inability to provide valid guarantees on their performance. Conformal prediction, a well-established methodology in the machine learning community, is an efficient and reliable strategy for providing trustworthy guarantees on the predictions of any black-box model under very minimal constraints. Bridging the gap between the safe machine learning and precision agriculture communities, this article showcases conformal prediction in action on the task of precision weeding through deep learning-based image classification. After a detailed presentation of the conformal prediction methodology and the development of a precision spraying pipeline based on a ''conformalized'' neural network and well-defined spraying decision rules, the article evaluates this pipeline on two real-world scenarios: one under in-distribution conditions, the other reflecting a near out-of-distribution setting. The results show that we are able to provide formal, i.e. certifiable, guarantees on spraying at least 90% of the weeds.
☆ Radial Distortion in Face Images: Detection and Impact
Acquiring face images of sufficiently high quality is important for online ID and travel document issuance applications using face recognition systems (FRS). Low-quality, manipulated (intentionally or unintentionally), or distorted images degrade the FRS performance and facilitate documents' misuse. Securing quality for enrolment images, especially in the unsupervised self-enrolment scenario via a smartphone, becomes important to assure FRS performance. In this work, we focus on the less studied area of radial distortion (a.k.a., the fish-eye effect) in face images and its impact on FRS performance. We introduce an effective radial distortion detection model that can detect and flag radial distortion in the enrolment scenario. We formalize the detection model as a face image quality assessment (FIQA) algorithm and provide a careful inspection of the effect of radial distortion on FRS performance. Evaluation results show excellent detection results for the proposed models, and the study on the impact on FRS uncovers valuable insights into how to best use these models in operational systems.
☆ BIOMEDICA: An Open Biomedical Image-Caption Archive, Dataset, and Vision-Language Models Derived from Scientific Literature
The development of vision-language models (VLMs) is driven by large-scale and diverse multimodal datasets. However, progress toward generalist biomedical VLMs is limited by the lack of annotated, publicly accessible datasets across biology and medicine. Existing efforts are restricted to narrow domains, missing the full diversity of biomedical knowledge encoded in scientific literature. To address this gap, we introduce BIOMEDICA, a scalable, open-source framework to extract, annotate, and serialize the entirety of the PubMed Central Open Access subset into an easy-to-use, publicly accessible dataset.Our framework produces a comprehensive archive with over 24 million unique image-text pairs from over 6 million articles. Metadata and expert-guided annotations are also provided. We demonstrate the utility and accessibility of our resource by releasing BMCA-CLIP, a suite of CLIP-style models continuously pre-trained on the BIOMEDICA dataset via streaming, eliminating the need to download 27 TB of data locally.On average, our models achieve state-of-the-art performance across 40 tasks - spanning pathology, radiology, ophthalmology, dermatology, surgery, molecular biology, parasitology, and cell biology - excelling in zero-shot classification with a 6.56% average improvement (as high as 29.8% and 17.5% in dermatology and ophthalmology, respectively), and stronger image-text retrieval, all while using 10x less compute. To foster reproducibility and collaboration, we release our codebase and dataset for the broader research community.
☆ Adaptive Noise-Tolerant Network for Image Segmentation
Unlike image classification and annotation, for which deep network models have achieved dominating superior performances compared to traditional computer vision algorithms, deep learning for automatic image segmentation still faces critical challenges. One of such hurdles is to obtain ground-truth segmentations as the training labels for deep network training. Especially when we study biomedical images, such as histopathological images (histo-images), it is unrealistic to ask for manual segmentation labels as the ground truth for training due to the fine image resolution as well as the large image size and complexity. In this paper, instead of relying on clean segmentation labels, we study whether and how integrating imperfect or noisy segmentation results from off-the-shelf segmentation algorithms may help achieve better segmentation results through a new Adaptive Noise-Tolerant Network (ANTN) model. We extend the noisy label deep learning to image segmentation with two novel aspects: (1) multiple noisy labels can be integrated into one deep learning model; (2) noisy segmentation modeling, including probabilistic parameters, is adaptive, depending on the given testing image appearance. Implementation of the new ANTN model on both the synthetic data and real-world histo-images demonstrates its effectiveness and superiority over off-the-shelf and other existing deep-learning-based image segmentation algorithms.
☆ Eye Sclera for Fair Face Image Quality Assessment
Fair operational systems are crucial in gaining and maintaining society's trust in face recognition systems (FRS). FRS start with capturing an image and assessing its quality before using it further for enrollment or verification. Fair Face Image Quality Assessment (FIQA) schemes therefore become equally important in the context of fair FRS. This work examines the sclera as a quality assessment region for obtaining a fair FIQA. The sclera region is agnostic to demographic variations and skin colour for assessing the quality of a face image. We analyze three skin tone related ISO/IEC face image quality assessment measures and assess the sclera region as an alternative area for assessing FIQ. Our analysis of the face dataset of individuals from different demographic groups representing different skin tones indicates sclera as an alternative to measure dynamic range, over- and under-exposure of face using sclera region alone. The sclera region being agnostic to skin tone, i.e., demographic factors, provides equal utility as a fair FIQA as shown by our Error-vs-Discard Characteristic (EDC) curve analysis.
☆ Robust Single Object Tracking in LiDAR Point Clouds under Adverse Weather Conditions
3D single object tracking (3DSOT) in LiDAR point clouds is a critical task for outdoor perception, enabling real-time perception of object location, orientation, and motion. Despite the impressive performance of current 3DSOT methods, evaluating them on clean datasets inadequately reflects their comprehensive performance, as the adverse weather conditions in real-world surroundings has not been considered. One of the main obstacles is the lack of adverse weather benchmarks for the evaluation of 3DSOT. To this end, this work proposes a challenging benchmark for LiDAR-based 3DSOT in adverse weather, which comprises two synthetic datasets (KITTI-A and nuScenes-A) and one real-world dataset (CADC-SOT) spanning three weather types: rain, fog, and snow. Based on this benchmark, five representative 3D trackers from different tracking frameworks conducted robustness evaluation, resulting in significant performance degradations. This prompts the question: What are the factors that cause current advanced methods to fail on such adverse weather samples? Consequently, we explore the impacts of adverse weather and answer the above question from three perspectives: 1) target distance; 2) template shape corruption; and 3) target shape corruption. Finally, based on domain randomization and contrastive learning, we designed a dual-branch tracking framework for adverse weather, named DRCT, achieving excellent performance in benchmarks.
comment: 14 pages
☆ MSV-Mamba: A Multiscale Vision Mamba Network for Echocardiography Segmentation
Ultrasound imaging frequently encounters challenges, such as those related to elevated noise levels, diminished spatiotemporal resolution, and the complexity of anatomical structures. These factors significantly hinder the model's ability to accurately capture and analyze structural relationships and dynamic patterns across various regions of the heart. Mamba, an emerging model, is one of the most cutting-edge approaches that is widely applied to diverse vision and language tasks. To this end, this paper introduces a U-shaped deep learning model incorporating a large-window Mamba scale (LMS) module and a hierarchical feature fusion approach for echocardiographic segmentation. First, a cascaded residual block serves as an encoder and is employed to incrementally extract multiscale detailed features. Second, a large-window multiscale mamba module is integrated into the decoder to capture global dependencies across regions and enhance the segmentation capability for complex anatomical structures. Furthermore, our model introduces auxiliary losses at each decoder layer and employs a dual attention mechanism to fuse multilayer features both spatially and across channels. This approach enhances segmentation performance and accuracy in delineating complex anatomical structures. Finally, the experimental results using the EchoNet-Dynamic and CAMUS datasets demonstrate that the model outperforms other methods in terms of both accuracy and robustness. For the segmentation of the left ventricular endocardium (${LV}_{endo}$), the model achieved optimal values of 95.01 and 93.36, respectively, while for the left ventricular epicardium (${LV}_{epi}$), values of 87.35 and 87.80, respectively, were achieved. This represents an improvement ranging between 0.54 and 1.11 compared with the best-performing model.
☆ Duplex: Dual Prototype Learning for Compositional Zero-Shot Learning
Compositional Zero-Shot Learning (CZSL) aims to enable models to recognize novel compositions of visual states and objects that were absent during training. Existing methods predominantly focus on learning semantic representations of seen compositions but often fail to disentangle the independent features of states and objects in images, thereby limiting their ability to generalize to unseen compositions. To address this challenge, we propose Duplex, a novel dual-prototype learning method that integrates semantic and visual prototypes through a carefully designed dual-branch architecture, enabling effective representation learning for compositional tasks. Duplex utilizes a Graph Neural Network (GNN) to adaptively update visual prototypes, capturing complex interactions between states and objects. Additionally, it leverages the strong visual-semantic alignment of pre-trained Vision-Language Models (VLMs) and employs a multi-path architecture combined with prompt engineering to align image and text representations, ensuring robust generalization. Extensive experiments on three benchmark datasets demonstrate that Duplex outperforms state-of-the-art methods in both closed-world and open-world settings.
☆ Matching Free Depth Recovery from Structured Light
We present a novel approach for depth estimation from images captured by structured light systems. Unlike many previous methods that rely on image matching process, our approach uses a density voxel grid to represent scene geometry, which is trained via self-supervised differentiable volume rendering. Our method leverages color fields derived from projected patterns in structured light systems during the rendering process, enabling the isolated optimization of the geometry field. This contributes to faster convergence and high-quality output. Additionally, we incorporate normalized device coordinates (NDC), a distortion loss, and a novel surface-based color loss to enhance geometric fidelity. Experimental results demonstrate that our method outperforms existing matching-based techniques in geometric performance for few-shot scenarios, achieving approximately a 60% reduction in average estimated depth errors on synthetic scenes and about 30% on real-world captured scenes. Furthermore, our approach delivers fast training, with a speed roughly three times faster than previous matching-free methods that employ implicit representations.
comment: 10 pages, 8 figures
☆ Dynamic Multimodal Fusion via Meta-Learning Towards Micro-Video Recommendation
Multimodal information (e.g., visual, acoustic, and textual) has been widely used to enhance representation learning for micro-video recommendation. For integrating multimodal information into a joint representation of micro-video, multimodal fusion plays a vital role in the existing micro-video recommendation approaches. However, the static multimodal fusion used in previous studies is insufficient to model the various relationships among multimodal information of different micro-videos. In this paper, we develop a novel meta-learning-based multimodal fusion framework called Meta Multimodal Fusion (MetaMMF), which dynamically assigns parameters to the multimodal fusion function for each micro-video during its representation learning. Specifically, MetaMMF regards the multimodal fusion of each micro-video as an independent task. Based on the meta information extracted from the multimodal features of the input task, MetaMMF parameterizes a neural network as the item-specific fusion function via a meta learner. We perform extensive experiments on three benchmark datasets, demonstrating the significant improvements over several state-of-the-art multimodal recommendation models, like MMGCN, LATTICE, and InvRL. Furthermore, we lighten our model by adopting canonical polyadic decomposition to improve the training efficiency, and validate its effectiveness through experimental results. Codes are available at https://github.com/hanliu95/MetaMMF.
comment: This paper has been accepted by ACM Transactions on Information Systems
☆ The Quest for Visual Understanding: A Journey Through the Evolution of Visual Question Answering
Visual Question Answering (VQA) is an interdisciplinary field that bridges the gap between computer vision (CV) and natural language processing(NLP), enabling Artificial Intelligence(AI) systems to answer questions about images. Since its inception in 2015, VQA has rapidly evolved, driven by advances in deep learning, attention mechanisms, and transformer-based models. This survey traces the journey of VQA from its early days, through major breakthroughs, such as attention mechanisms, compositional reasoning, and the rise of vision-language pre-training methods. We highlight key models, datasets, and techniques that shaped the development of VQA systems, emphasizing the pivotal role of transformer architectures and multimodal pre-training in driving recent progress. Additionally, we explore specialized applications of VQA in domains like healthcare and discuss ongoing challenges, such as dataset bias, model interpretability, and the need for common-sense reasoning. Lastly, we discuss the emerging trends in large multimodal language models and the integration of external knowledge, offering insights into the future directions of VQA. This paper aims to provide a comprehensive overview of the evolution of VQA, highlighting both its current state and potential advancements.
☆ RMAvatar: Photorealistic Human Avatar Reconstruction from Monocular Video Based on Rectified Mesh-embedded Gaussians
We introduce RMAvatar, a novel human avatar representation with Gaussian splatting embedded on mesh to learn clothed avatar from a monocular video. We utilize the explicit mesh geometry to represent motion and shape of a virtual human and implicit appearance rendering with Gaussian Splatting. Our method consists of two main modules: Gaussian initialization module and Gaussian rectification module. We embed Gaussians into triangular faces and control their motion through the mesh, which ensures low-frequency motion and surface deformation of the avatar. Due to the limitations of LBS formula, the human skeleton is hard to control complex non-rigid transformations. We then design a pose-related Gaussian rectification module to learn fine-detailed non-rigid deformations, further improving the realism and expressiveness of the avatar. We conduct extensive experiments on public datasets, RMAvatar shows state-of-the-art performance on both rendering quality and quantitative evaluations. Please see our project page at https://rm-avatar.github.io.
comment: CVM2025
☆ Dual Scale-aware Adaptive Masked Knowledge Distillation for Object Detection
Recent feature masking knowledge distillation methods make use of attention mechanisms to identify either important spatial regions or channel clues for discriminative feature reconstruction. However, most of existing strategies perform global attention-guided feature masking distillation without delving into fine-grained visual clues in feature maps. In particular, uncovering locality-aware clues across different scales are conducive to reconstructing region-aware features, thereby significantly benefiting distillation performance. In this study, we propose a fine-grained adaptive feature masking distillation framework for accurate object detection. Different from previous methods in which global masking is performed on single-scale feature maps, we explore the scale-aware feature masking by performing feature distillation across various scales, such that the object-aware locality is encoded for improved feature reconstruction. In addition, our fine-grained feature distillation strategy is combined with a masking logits distillation scheme in which logits difference between teacher and student networks is utilized to guide the distillation process. Thus, it can help the student model to better learn from the teacher counterpart with improved knowledge transfer. Extensive experiments for detection task demonstrate the superiority of our method. For example, when RetinaNet, RepPoints and Cascade Mask RCNN are used as teacher detectors, the student network achieves mAP scores of 41.5\%, 42.9\%, and 42.6\%, respectively, outperforming state-of-the-art methods such as DMKD and FreeKD.
☆ Collaborative Learning for 3D Hand-Object Reconstruction and Compositional Action Recognition from Egocentric RGB Videos Using Superquadrics AAAI 2025
With the availability of egocentric 3D hand-object interaction datasets, there is increasing interest in developing unified models for hand-object pose estimation and action recognition. However, existing methods still struggle to recognise seen actions on unseen objects due to the limitations in representing object shape and movement using 3D bounding boxes. Additionally, the reliance on object templates at test time limits their generalisability to unseen objects. To address these challenges, we propose to leverage superquadrics as an alternative 3D object representation to bounding boxes and demonstrate their effectiveness on both template-free object reconstruction and action recognition tasks. Moreover, as we find that pure appearance-based methods can outperform the unified methods, the potential benefits from 3D geometric information remain unclear. Therefore, we study the compositionality of actions by considering a more challenging task where the training combinations of verbs and nouns do not overlap with the testing split. We extend H2O and FPHA datasets with compositional splits and design a novel collaborative learning framework that can explicitly reason about the geometric relations between hands and the manipulated object. Through extensive quantitative and qualitative evaluations, we demonstrate significant improvements over the state-of-the-arts in (compositional) action recognition.
comment: Accepted to AAAI 2025
☆ Video Quality Assessment for Online Processing: From Spatial to Temporal Sampling
With the rapid development of multimedia processing and deep learning technologies, especially in the field of video understanding, video quality assessment (VQA) has achieved significant progress. Although researchers have moved from designing efficient video quality mapping models to various research directions, in-depth exploration of the effectiveness-efficiency trade-offs of spatio-temporal modeling in VQA models is still less sufficient. Considering the fact that videos have highly redundant information, this paper investigates this problem from the perspective of joint spatial and temporal sampling, aiming to seek the answer to how little information we should keep at least when feeding videos into the VQA models while with acceptable performance sacrifice. To this end, we drastically sample the video's information from both spatial and temporal dimensions, and the heavily squeezed video is then fed into a stable VQA model. Comprehensive experiments regarding joint spatial and temporal sampling are conducted on six public video quality databases, and the results demonstrate the acceptable performance of the VQA model when throwing away most of the video information. Furthermore, with the proposed joint spatial and temporal sampling strategy, we make an initial attempt to design an online VQA model, which is instantiated by as simple as possible a spatial feature extractor, a temporal feature fusion module, and a global quality regression module. Through quantitative and qualitative experiments, we verify the feasibility of online VQA model by simplifying itself and reducing input.
☆ Representation Learning of Point Cloud Upsampling in Global and Local Inputs
In recent years, point cloud upsampling has been widely applied in fields such as 3D reconstruction. Our study investigates the factors influencing point cloud upsampling on both global and local levels through representation learning. Specifically, the paper inputs global and local information of the same point cloud model object into two encoders to extract these features, fuses them, and then feeds the combined features into an upsampling decoder. The goal is to address issues of sparsity and noise in point clouds by leveraging prior knowledge from both global and local inputs. And the proposed framework can be applied to any state-of-the-art point cloud upsampling neural network. Experiments were conducted on a series of autoencoder-based models utilizing deep learning, yielding interpretability for both global and local inputs, and it has been proven in the results that our proposed framework can further improve the upsampling effect in previous SOTA works. At the same time, the Saliency Map reflects the differences between global and local feature inputs, as well as the effectiveness of training with both inputs in parallel.
☆ Label Calibration in Source Free Domain Adaptation WACV
Source-free domain adaptation (SFDA) utilizes a pre-trained source model with unlabeled target data. Self-supervised SFDA techniques generate pseudolabels from the pre-trained source model, but these pseudolabels often contain noise due to domain discrepancies between the source and target domains. Traditional self-supervised SFDA techniques rely on deterministic model predictions using the softmax function, leading to unreliable pseudolabels. In this work, we propose to introduce predictive uncertainty and softmax calibration for pseudolabel refinement using evidential deep learning. The Dirichlet prior is placed over the output of the target network to capture uncertainty using evidence with a single forward pass. Furthermore, softmax calibration solves the translation invariance problem to assist in learning with noisy labels. We incorporate a combination of evidential deep learning loss and information maximization loss with calibrated softmax in both prior and non-prior target knowledge SFDA settings. Extensive experimental analysis shows that our method outperforms other state-of-the-art methods on benchmark datasets.
comment: Accepted in IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2025
☆ Enhancing Image Generation Fidelity via Progressive Prompts ICASSP 2025
The diffusion transformer (DiT) architecture has attracted significant attention in image generation, achieving better fidelity, performance, and diversity. However, most existing DiT - based image generation methods focus on global - aware synthesis, and regional prompt control has been less explored. In this paper, we propose a coarse - to - fine generation pipeline for regional prompt - following generation. Specifically, we first utilize the powerful large language model (LLM) to generate both high - level descriptions of the image (such as content, topic, and objects) and low - level descriptions (such as details and style). Then, we explore the influence of cross - attention layers at different depths. We find that deeper layers are always responsible for high - level content control, while shallow layers handle low - level content control. Various prompts are injected into the proposed regional cross - attention control for coarse - to - fine generation. By using the proposed pipeline, we enhance the controllability of DiT - based image generation. Extensive quantitative and qualitative results show that our pipeline can improve the performance of the generated images.
comment: Accepted by ICASSP 2025, Github: https://github.com/ZhenXiong-dl/ICASSP2025-RCAC
☆ Hierarchical Superpixel Segmentation via Structural Information Theory SDM 2025
Superpixel segmentation is a foundation for many higher-level computer vision tasks, such as image segmentation, object recognition, and scene understanding. Existing graph-based superpixel segmentation methods typically concentrate on the relationships between a given pixel and its directly adjacent pixels while overlooking the influence of non-adjacent pixels. These approaches do not fully leverage the global information in the graph, leading to suboptimal segmentation quality. To address this limitation, we present SIT-HSS, a hierarchical superpixel segmentation method based on structural information theory. Specifically, we first design a novel graph construction strategy that incrementally explores the pixel neighborhood to add edges based on 1-dimensional structural entropy (1D SE). This strategy maximizes the retention of graph information while avoiding an overly complex graph structure. Then, we design a new 2D SE-guided hierarchical graph partitioning method, which iteratively merges pixel clusters layer by layer to reduce the graph's 2D SE until a predefined segmentation scale is achieved. Experimental results on three benchmark datasets demonstrate that the SIT-HSS performs better than state-of-the-art unsupervised superpixel segmentation algorithms. The source code is available at \url{https://github.com/SELGroup/SIT-HSS}.
comment: Accepted by SDM 2025
☆ SFC-GAN: A Generative Adversarial Network for Brain Functional and Structural Connectome Translation
Modern brain imaging technologies have enabled the detailed reconstruction of human brain connectomes, capturing structural connectivity (SC) from diffusion MRI and functional connectivity (FC) from functional MRI. Understanding the intricate relationships between SC and FC is vital for gaining deeper insights into the brain's functional and organizational mechanisms. However, obtaining both SC and FC modalities simultaneously remains challenging, hindering comprehensive analyses. Existing deep generative models typically focus on synthesizing a single modality or unidirectional translation between FC and SC, thereby missing the potential benefits of bi-directional translation, especially in scenarios where only one connectome is available. Therefore, we propose Structural-Functional Connectivity GAN (SFC-GAN), a novel framework for bidirectional translation between SC and FC. This approach leverages the CycleGAN architecture, incorporating convolutional layers to effectively capture the spatial structures of brain connectomes. To preserve the topological integrity of these connectomes, we employ a structure-preserving loss that guides the model in capturing both global and local connectome patterns while maintaining symmetry. Our framework demonstrates superior performance in translating between SC and FC, outperforming baseline models in similarity and graph property evaluations compared to ground truth data, each translated modality can be effectively utilized for downstream classification.
comment: 5 pages, 2 figures
☆ Protego: Detecting Adversarial Examples for Vision Transformers via Intrinsic Capabilities
Transformer models have excelled in natural language tasks, prompting the vision community to explore their implementation in computer vision problems. However, these models are still influenced by adversarial examples. In this paper, we investigate the attack capabilities of six common adversarial attacks on three pretrained ViT models to reveal the vulnerability of ViT models. To understand and analyse the bias in neural network decisions when the input is adversarial, we use two visualisation techniques that are attention rollout and grad attention rollout. To prevent ViT models from adversarial attack, we propose Protego, a detection framework that leverages the transformer intrinsic capabilities to detection adversarial examples of ViT models. Nonetheless, this is challenging due to a diversity of attack strategies that may be adopted by adversaries. Inspired by the attention mechanism, we know that the token of prediction contains all the information from the input sample. Additionally, the attention region for adversarial examples differs from that of normal examples. Given these points, we can train a detector that achieves superior performance than existing detection methods to identify adversarial examples. Our experiments have demonstrated the high effectiveness of our detection method. For these six adversarial attack methods, our detector's AUC scores all exceed 0.95. Protego may advance investigations in metaverse security.
comment: Accepted by IEEE MetaCom 2024
☆ Rethinking Knowledge in Distillation: An In-context Sample Retrieval Perspective
Conventional knowledge distillation (KD) approaches are designed for the student model to predict similar output as the teacher model for each sample. Unfortunately, the relationship across samples with same class is often neglected. In this paper, we explore to redefine the knowledge in distillation, capturing the relationship between each sample and its corresponding in-context samples (a group of similar samples with the same or different classes), and perform KD from an in-context sample retrieval perspective. As KD is a type of learned label smoothing regularization (LSR), we first conduct a theoretical analysis showing that the teacher's knowledge from the in-context samples is a crucial contributor to regularize the student training with the corresponding samples. Buttressed by the analysis, we propose a novel in-context knowledge distillation (IC-KD) framework that shows its superiority across diverse KD paradigms (offline, online, and teacher-free KD). Firstly, we construct a feature memory bank from the teacher model and retrieve in-context samples for each corresponding sample through retrieval-based learning. We then introduce Positive In-Context Distillation (PICD) to reduce the discrepancy between a sample from the student and the aggregated in-context samples with the same class from the teacher in the logit space. Moreover, Negative In-Context Distillation (NICD) is introduced to separate a sample from the student and the in-context samples with different classes from the teacher in the logit space. Extensive experiments demonstrate that IC-KD is effective across various types of KD, and consistently achieves state-of-the-art performance on CIFAR-100 and ImageNet datasets.
☆ IoT-Based Real-Time Medical-Related Human Activity Recognition Using Skeletons and Multi-Stage Deep Learning for Healthcare
The Internet of Things (IoT) and mobile technology have significantly transformed healthcare by enabling real-time monitoring and diagnosis of patients. Recognizing medical-related human activities (MRHA) is pivotal for healthcare systems, particularly for identifying actions that are critical to patient well-being. However, challenges such as high computational demands, low accuracy, and limited adaptability persist in Human Motion Recognition (HMR). While some studies have integrated HMR with IoT for real-time healthcare applications, limited research has focused on recognizing MRHA as essential for effective patient monitoring. This study proposes a novel HMR method for MRHA detection, leveraging multi-stage deep learning techniques integrated with IoT. The approach employs EfficientNet to extract optimized spatial features from skeleton frame sequences using seven Mobile Inverted Bottleneck Convolutions (MBConv) blocks, followed by ConvLSTM to capture spatio-temporal patterns. A classification module with global average pooling, a fully connected layer, and a dropout layer generates the final predictions. The model is evaluated on the NTU RGB+D 120 and HMDB51 datasets, focusing on MRHA, such as sneezing, falling, walking, sitting, etc. It achieves 94.85% accuracy for cross-subject evaluations and 96.45% for cross-view evaluations on NTU RGB+D 120, along with 89.00% accuracy on HMDB51. Additionally, the system integrates IoT capabilities using a Raspberry Pi and GSM module, delivering real-time alerts via Twilios SMS service to caregivers and patients. This scalable and efficient solution bridges the gap between HMR and IoT, advancing patient monitoring, improving healthcare outcomes, and reducing costs.
☆ Detection of AI Deepfake and Fraud in Online Payments Using GAN-Based Models
This study explores the use of Generative Adversarial Networks (GANs) to detect AI deepfakes and fraudulent activities in online payment systems. With the growing prevalence of deepfake technology, which can manipulate facial features in images and videos, the potential for fraud in online transactions has escalated. Traditional security systems struggle to identify these sophisticated forms of fraud. This research proposes a novel GAN-based model that enhances online payment security by identifying subtle manipulations in payment images. The model is trained on a dataset consisting of real-world online payment images and deepfake images generated using advanced GAN architectures, such as StyleGAN and DeepFake. The results demonstrate that the proposed model can accurately distinguish between legitimate transactions and deepfakes, achieving a high detection rate above 95%. This approach significantly improves the robustness of payment systems against AI-driven fraud. The paper contributes to the growing field of digital security, offering insights into the application of GANs for fraud detection in financial services. Keywords- Payment Security, Image Recognition, Generative Adversarial Networks, AI Deepfake, Fraudulent Activities
comment: The paper will be published and indexed by IEEE at 2025 8th International Conference on Advanced Algorithms and Control Engineering (ICAACE 2025)
☆ UNetVL: Enhancing 3D Medical Image Segmentation with Chebyshev KAN Powered Vision-LSTM
3D medical image segmentation has progressed considerably due to Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), yet these methods struggle to balance long-range dependency acquisition with computational efficiency. To address this challenge, we propose UNETVL (U-Net Vision-LSTM), a novel architecture that leverages recent advancements in temporal information processing. UNETVL incorporates Vision-LSTM (ViL) for improved scalability and memory functions, alongside an efficient Chebyshev Kolmogorov-Arnold Networks (KAN) to handle complex and long-range dependency patterns more effectively. We validated our method on the ACDC and AMOS2022 (post challenge Task 2) benchmark datasets, showing a significant improvement in mean Dice score compared to recent state-of-the-art approaches, especially over its predecessor, UNETR, with increases of 7.3% on ACDC and 15.6% on AMOS, respectively. Extensive ablation studies were conducted to demonstrate the impact of each component in UNETVL, providing a comprehensive understanding of its architecture. Our code is available at https://github.com/tgrex6/UNETVL, facilitating further research and applications in this domain.
☆ A Multi-Modal Deep Learning Framework for Pan-Cancer Prognosis
Prognostic task is of great importance as it closely related to the survival analysis of patients, the optimization of treatment plans and the allocation of resources. The existing prognostic models have shown promising results on specific datasets, but there are limitations in two aspects. On the one hand, they merely explore certain types of modal data, such as patient histopathology WSI and gene expression analysis. On the other hand, they adopt the per-cancer-per-model paradigm, which means the trained models can only predict the prognostic effect of a single type of cancer, resulting in weak generalization ability. In this paper, a deep-learning based model, named UMPSNet, is proposed. Specifically, to comprehensively understand the condition of patients, in addition to constructing encoders for histopathology images and genomic expression profiles respectively, UMPSNet further integrates four types of important meta data (demographic information, cancer type information, treatment protocols, and diagnosis results) into text templates, and then introduces a text encoder to extract textual features. In addition, the optimal transport OT-based attention mechanism is utilized to align and fuse features of different modalities. Furthermore, a guided soft mixture of experts (GMoE) mechanism is introduced to effectively address the issue of distribution differences among multiple cancer datasets. By incorporating the multi-modality of patient data and joint training, UMPSNet outperforms all SOTA approaches, and moreover, it demonstrates the effectiveness and generalization ability of the proposed learning paradigm of a single model for multiple cancer types. The code of UMPSNet is available at https://github.com/binging512/UMPSNet.
☆ SplatMAP: Online Dense Monocular SLAM with 3D Gaussian Splatting
Achieving high-fidelity 3D reconstruction from monocular video remains challenging due to the inherent limitations of traditional methods like Structure-from-Motion (SfM) and monocular SLAM in accurately capturing scene details. While differentiable rendering techniques such as Neural Radiance Fields (NeRF) address some of these challenges, their high computational costs make them unsuitable for real-time applications. Additionally, existing 3D Gaussian Splatting (3DGS) methods often focus on photometric consistency, neglecting geometric accuracy and failing to exploit SLAM's dynamic depth and pose updates for scene refinement. We propose a framework integrating dense SLAM with 3DGS for real-time, high-fidelity dense reconstruction. Our approach introduces SLAM-Informed Adaptive Densification, which dynamically updates and densifies the Gaussian model by leveraging dense point clouds from SLAM. Additionally, we incorporate Geometry-Guided Optimization, which combines edge-aware geometric constraints and photometric consistency to jointly optimize the appearance and geometry of the 3DGS scene representation, enabling detailed and accurate SLAM mapping reconstruction. Experiments on the Replica and TUM-RGBD datasets demonstrate the effectiveness of our approach, achieving state-of-the-art results among monocular systems. Specifically, our method achieves a PSNR of 36.864, SSIM of 0.985, and LPIPS of 0.040 on Replica, representing improvements of 10.7%, 6.4%, and 49.4%, respectively, over the previous SOTA. On TUM-RGBD, our method outperforms the closest baseline by 10.2%, 6.6%, and 34.7% in the same metrics. These results highlight the potential of our framework in bridging the gap between photometric and geometric dense 3D scene representations, paving the way for practical and efficient monocular dense reconstruction.
☆ LEO: Boosting Mixture of Vision Encoders for Multimodal Large Language Models
Enhanced visual understanding serves as a cornerstone for multimodal large language models (MLLMs). Recent hybrid MLLMs incorporate a mixture of vision experts to address the limitations of using a single vision encoder and excessively long visual tokens. Despite the progress of these MLLMs, a research gap remains in effectively integrating diverse vision encoders. This work explores fusion strategies of visual tokens for hybrid MLLMs, leading to the design of LEO, a novel MLLM with a dual-branch vision encoder framework that incorporates a post-adaptation fusion strategy and adaptive tiling: for each segmented tile of the input images, LEO sequentially interleaves the visual tokens from its two vision encoders. Extensive evaluation across 13 vision-language benchmarks reveals that LEO outperforms state-of-the-art open-source MLLMs and hybrid MLLMs on the majority of tasks. Furthermore, we show that LEO can be adapted to the specialized domain of autonomous driving without altering the model architecture or training recipe, achieving competitive performance compared to existing baselines. The code and model will be publicly available.
☆ Universal Training of Neural Networks to Achieve Bayes Optimal Classification Accuracy ICASSP 2025
This work invokes the notion of $f$-divergence to introduce a novel upper bound on the Bayes error rate of a general classification task. We show that the proposed bound can be computed by sampling from the output of a parameterized model. Using this practical interpretation, we introduce the Bayes optimal learning threshold (BOLT) loss whose minimization enforces a classification model to achieve the Bayes error rate. We validate the proposed loss for image and text classification tasks, considering MNIST, Fashion-MNIST, CIFAR-10, and IMDb datasets. Numerical experiments demonstrate that models trained with BOLT achieve performance on par with or exceeding that of cross-entropy, particularly on challenging datasets. This highlights the potential of BOLT in improving generalization.
comment: Accepted to ICASSP 2025
☆ Boosting Sclera Segmentation through Semi-supervised Learning with Fewer Labels
Sclera segmentation is crucial for developing automatic eye-related medical computer-aided diagnostic systems, as well as for personal identification and verification, because the sclera contains distinct personal features. Deep learning-based sclera segmentation has achieved significant success compared to traditional methods that rely on hand-crafted features, primarily because it can autonomously extract critical output-related features without the need to consider potential physical constraints. However, achieving accurate sclera segmentation using these methods is challenging due to the scarcity of high-quality, fully labeled datasets, which depend on costly, labor-intensive medical acquisition and expertise. To address this challenge, this paper introduces a novel sclera segmentation framework that excels with limited labeled samples. Specifically, we employ a semi-supervised learning method that integrates domain-specific improvements and image-based spatial transformations to enhance segmentation performance. Additionally, we have developed a real-world eye diagnosis dataset to enrich the evaluation process. Extensive experiments on our dataset and two additional public datasets demonstrate the effectiveness and superiority of our proposed method, especially with significantly fewer labeled samples.
comment: Under review, 19 pages, 9 figures, 4 tables
☆ A Heterogeneous Multimodal Graph Learning Framework for Recognizing User Emotions in Social Networks
The rapid expansion of social media platforms has provided unprecedented access to massive amounts of multimodal user-generated content. Comprehending user emotions can provide valuable insights for improving communication and understanding of human behaviors. Despite significant advancements in Affective Computing, the diverse factors influencing user emotions in social networks remain relatively understudied. Moreover, there is a notable lack of deep learning-based methods for predicting user emotions in social networks, which could be addressed by leveraging the extensive multimodal data available. This work presents a novel formulation of personalized emotion prediction in social networks based on heterogeneous graph learning. Building upon this formulation, we design HMG-Emo, a Heterogeneous Multimodal Graph Learning Framework that utilizes deep learning-based features for user emotion recognition. Additionally, we include a dynamic context fusion module in HMG-Emo that is capable of adaptively integrating the different modalities in social media data. Through extensive experiments, we demonstrate the effectiveness of HMG-Emo and verify the superiority of adopting a graph neural network-based approach, which outperforms existing baselines that use rich hand-crafted features. To the best of our knowledge, HMG-Emo is the first multimodal and deep-learning-based approach to predict personalized emotions within online social networks. Our work highlights the significance of exploiting advanced deep learning techniques for less-explored problems in Affective Computing.
☆ Fixing the Scale and Shift in Monocular Depth For Camera Pose Estimation
Recent advances in monocular depth prediction have led to significantly improved depth prediction accuracy. In turn, this enables various applications to use such depth predictions. In this paper, we propose a novel framework for estimating the relative pose between two cameras from point correspondences with associated monocular depths. Since depth predictions are typically defined up to an unknown scale and shift parameter, our solvers jointly estimate both scale and shift parameters together with the camera pose. We derive efficient solvers for three cases: (1) two calibrated cameras, (2) two uncalibrated cameras with an unknown but shared focal length, and (3) two uncalibrated cameras with unknown and different focal lengths. Experiments on synthetic and real data, including experiments with depth maps estimated by 11 different depth predictors, show the practical viability of our solvers. Compared to prior work, our solvers achieve state-of-the-art results on two large-scale, real-world datasets. The source code is available at https://github.com/yaqding/pose_monodepth
comment: 14 pages
☆ Democratizing Text-to-Image Masked Generative Models with Compact Text-Aware One-Dimensional Tokens
Image tokenizers form the foundation of modern text-to-image generative models but are notoriously difficult to train. Furthermore, most existing text-to-image models rely on large-scale, high-quality private datasets, making them challenging to replicate. In this work, we introduce Text-Aware Transformer-based 1-Dimensional Tokenizer (TA-TiTok), an efficient and powerful image tokenizer that can utilize either discrete or continuous 1-dimensional tokens. TA-TiTok uniquely integrates textual information during the tokenizer decoding stage (i.e., de-tokenization), accelerating convergence and enhancing performance. TA-TiTok also benefits from a simplified, yet effective, one-stage training process, eliminating the need for the complex two-stage distillation used in previous 1-dimensional tokenizers. This design allows for seamless scalability to large datasets. Building on this, we introduce a family of text-to-image Masked Generative Models (MaskGen), trained exclusively on open data while achieving comparable performance to models trained on private data. We aim to release both the efficient, strong TA-TiTok tokenizers and the open-data, open-weight MaskGen models to promote broader access and democratize the field of text-to-image masked generative models.
comment: Project page at https://tacju.github.io/projects/maskgen.html
☆ Testing Human-Hand Segmentation on In-Distribution and Out-of-Distribution Data in Human-Robot Interactions Using a Deep Ensemble Model
Reliable detection and segmentation of human hands are critical for enhancing safety and facilitating advanced interactions in human-robot collaboration. Current research predominantly evaluates hand segmentation under in-distribution (ID) data, which reflects the training data of deep learning (DL) models. However, this approach fails to address out-of-distribution (OOD) scenarios that often arise in real-world human-robot interactions. In this study, we present a novel approach by evaluating the performance of pre-trained DL models under both ID data and more challenging OOD scenarios. To mimic realistic industrial scenarios, we designed a diverse dataset featuring simple and cluttered backgrounds with industrial tools, varying numbers of hands (0 to 4), and hands with and without gloves. For OOD scenarios, we incorporated unique and rare conditions such as finger-crossing gestures and motion blur from fast-moving hands, addressing both epistemic and aleatoric uncertainties. To ensure multiple point of views (PoVs), we utilized both egocentric cameras, mounted on the operator's head, and static cameras to capture RGB images of human-robot interactions. This approach allowed us to account for multiple camera perspectives while also evaluating the performance of models trained on existing egocentric datasets as well as static-camera datasets. For segmentation, we used a deep ensemble model composed of UNet and RefineNet as base learners. Performance evaluation was conducted using segmentation metrics and uncertainty quantification via predictive entropy. Results revealed that models trained on industrial datasets outperformed those trained on non-industrial datasets, highlighting the importance of context-specific training. Although all models struggled with OOD scenarios, those trained on industrial datasets demonstrated significantly better generalization.
☆ Pedestrian Trajectory Prediction Based on Social Interactions Learning With Random Weights
Pedestrian trajectory prediction is a critical technology in the evolution of self-driving cars toward complete artificial intelligence. Over recent years, focusing on the trajectories of pedestrians to model their social interactions has surged with great interest in more accurate trajectory predictions. However, existing methods for modeling pedestrian social interactions rely on pre-defined rules, struggling to capture non-explicit social interactions. In this work, we propose a novel framework named DTGAN, which extends the application of Generative Adversarial Networks (GANs) to graph sequence data, with the primary objective of automatically capturing implicit social interactions and achieving precise predictions of pedestrian trajectory. DTGAN innovatively incorporates random weights within each graph to eliminate the need for pre-defined interaction rules. We further enhance the performance of DTGAN by exploring diverse task loss functions during adversarial training, which yields improvements of 16.7\% and 39.3\% on metrics ADE and FDE, respectively. The effectiveness and accuracy of our framework are verified on two public datasets. The experimental results show that our proposed DTGAN achieves superior performance and is well able to understand pedestrians' intentions.
comment: 13 pages,7 figures,Accepted to IEEE Transactions on Multimedia (TMM)
☆ C2PD: Continuity-Constrained Pixelwise Deformation for Guided Depth Super-Resolution
Guided depth super-resolution (GDSR) has demonstrated impressive performance across a wide range of domains, with numerous methods being proposed. However, existing methods often treat depth maps as images, where shading values are computed discretely, making them struggle to effectively restore the continuity inherent in the depth map. In this paper, we propose a novel approach that maximizes the utilization of spatial characteristics in depth, coupled with human abstract perception of real-world substance, by transforming the GDSR issue into deformation of a roughcast with ideal plasticity, which can be deformed by force like a continuous object. Specifically, we firstly designed a cross-modal operation, Continuity-constrained Asymmetrical Pixelwise Operation (CAPO), which can mimic the process of deforming an isovolumetrically flexible object through external forces. Utilizing CAPO as the fundamental component, we develop the Pixelwise Cross Gradient Deformation (PCGD), which is capable of emulating operations on ideal plastic objects (without volume constraint). Notably, our approach demonstrates state-of-the-art performance across four widely adopted benchmarks for GDSR, with significant advantages in large-scale tasks and generalizability.
Dataset Distillation as Pushforward Optimal Quantization
Dataset distillation aims to find a synthetic training set such that training on the synthetic data achieves similar performance to training on real data, with orders of magnitude less computational requirements. Existing methods can be broadly categorized as either bi-level optimization problems that have neural network training heuristics as the lower level problem, or disentangled methods that bypass the bi-level optimization by matching distributions of data. The latter method has the major advantages of speed and scalability in terms of size of both training and distilled datasets. We demonstrate that when equipped with an encoder-decoder structure, the empirically successful disentangled methods can be reformulated as an optimal quantization problem, where a finite set of points is found to approximate the underlying probability measure by minimizing the expected projection distance. In particular, we link existing disentangled dataset distillation methods to the classical optimal quantization and Wasserstein barycenter problems, demonstrating consistency of distilled datasets for diffusion-based generative priors. We propose a simple extension of the state-of-the-art data distillation method D4M, achieving better performance on the ImageNet-1K dataset with trivial additional computation, and state-of-the-art performance in higher image-per-class settings.
☆ BlobGEN-Vid: Compositional Text-to-Video Generation with Blob Video Representations
Existing video generation models struggle to follow complex text prompts and synthesize multiple objects, raising the need for additional grounding input for improved controllability. In this work, we propose to decompose videos into visual primitives - blob video representation, a general representation for controllable video generation. Based on blob conditions, we develop a blob-grounded video diffusion model named BlobGEN-Vid that allows users to control object motions and fine-grained object appearance. In particular, we introduce a masked 3D attention module that effectively improves regional consistency across frames. In addition, we introduce a learnable module to interpolate text embeddings so that users can control semantics in specific frames and obtain smooth object transitions. We show that our framework is model-agnostic and build BlobGEN-Vid based on both U-Net and DiT-based video diffusion models. Extensive experimental results show that BlobGEN-Vid achieves superior zero-shot video generation ability and state-of-the-art layout controllability on multiple benchmarks. When combined with an LLM for layout planning, our framework even outperforms proprietary text-to-video generators in terms of compositional accuracy.
comment: Project page: https://blobgen-vid2.github.io/
♻ ☆ The Sound of Water: Inferring Physical Properties from Pouring Liquids ICASSP 2025
We study the connection between audio-visual observations and the underlying physics of a mundane yet intriguing everyday activity: pouring liquids. Given only the sound of liquid pouring into a container, our objective is to automatically infer physical properties such as the liquid level, the shape and size of the container, the pouring rate and the time to fill. To this end, we: (i) show in theory that these properties can be determined from the fundamental frequency (pitch); (ii) train a pitch detection model with supervision from simulated data and visual data with a physics-inspired objective; (iii) introduce a new large dataset of real pouring videos for a systematic study; (iv) show that the trained model can indeed infer these physical properties for real data; and finally, (v) we demonstrate strong generalization to various container shapes, other datasets, and in-the-wild YouTube videos. Our work presents a keen understanding of a narrow yet rich problem at the intersection of acoustics, physics, and learning. It opens up applications to enhance multisensory perception in robotic pouring.
comment: Project page at https://bpiyush.github.io/pouring-water-website. Short version accepted to ICASSP 2025
♻ ☆ Robot Synesthesia: A Sound and Emotion Guided AI Painter
If a picture paints a thousand words, sound may voice a million. While recent robotic painting and image synthesis methods have achieved progress in generating visuals from text inputs, the translation of sound into images is vastly unexplored. Generally, sound-based interfaces and sonic interactions have the potential to expand accessibility and control for the user and provide a means to convey complex emotions and the dynamic aspects of the real world. In this paper, we propose an approach for using sound and speech to guide a robotic painting process, known here as robot synesthesia. For general sound, we encode the simulated paintings and input sounds into the same latent space. For speech, we decouple speech into its transcribed text and the tone of the speech. Whereas we use the text to control the content, we estimate the emotions from the tone to guide the mood of the painting. Our approach has been fully integrated with FRIDA, a robotic painting framework, adding sound and speech to FRIDA's existing input modalities, such as text and style. In two surveys, participants were able to correctly guess the emotion or natural sound used to generate a given painting more than twice as likely as random chance. On our sound-guided image manipulation and music-guided paintings, we discuss the results qualitatively.
comment: 9 pages, 10 figures
♻ ☆ Quilt-1M: One Million Image-Text Pairs for Histopathology
Recent accelerations in multi-modal applications have been made possible with the plethora of image and text data available online. However, the scarcity of analogous data in the medical field, specifically in histopathology, has slowed comparable progress. To enable similar representation learning for histopathology, we turn to YouTube, an untapped resource of videos, offering $1,087$ hours of valuable educational histopathology videos from expert clinicians. From YouTube, we curate QUILT: a large-scale vision-language dataset consisting of $802, 144$ image and text pairs. QUILT was automatically curated using a mixture of models, including large language models, handcrafted algorithms, human knowledge databases, and automatic speech recognition. In comparison, the most comprehensive datasets curated for histopathology amass only around $200$K samples. We combine QUILT with datasets from other sources, including Twitter, research papers, and the internet in general, to create an even larger dataset: QUILT-1M, with $1$M paired image-text samples, marking it as the largest vision-language histopathology dataset to date. We demonstrate the value of QUILT-1M by fine-tuning a pre-trained CLIP model. Our model outperforms state-of-the-art models on both zero-shot and linear probing tasks for classifying new histopathology images across $13$ diverse patch-level datasets of $8$ different sub-pathologies and cross-modal retrieval tasks.
♻ ☆ Enhance Eye Disease Detection using Learnable Probabilistic Discrete Latents in Machine Learning Architectures
Ocular diseases, including diabetic retinopathy and glaucoma, present a significant public health challenge due to their high prevalence and potential for causing vision impairment. Early and accurate diagnosis is crucial for effective treatment and management. In recent years, deep learning models have emerged as powerful tools for analysing medical images, such as retina imaging. However, challenges persist in model relibability and uncertainty estimation, which are critical for clinical decision-making. This study leverages the probabilistic framework of Generative Flow Networks (GFlowNets) to learn the posterior distribution over latent discrete dropout masks for the classification and analysis of ocular diseases using fundus images. We develop a robust and generalizable method that utilizes GFlowOut integrated with ResNet18 and ViT models as the backbone in identifying various ocular conditions. This study employs a unique set of dropout masks - none, random, bottomup, and topdown - to enhance model performance in analyzing these fundus images. Our results demonstrate that our learnable probablistic latents significantly improves accuracy, outperforming the traditional dropout approach. We utilize a gradient map calculation method, Grad-CAM, to assess model explainability, observing that the model accurately focuses on critical image regions for predictions. The integration of GFlowOut in neural networks presents a promising advancement in the automated diagnosis of ocular diseases, with implications for improving clinical workflows and patient outcomes.
♻ ☆ RGB-D Indiscernible Object Counting in Underwater Scenes
Recently, indiscernible/camouflaged scene understanding has attracted lots of research attention in the vision community. We further advance the frontier of this field by systematically studying a new challenge named indiscernible object counting (IOC), the goal of which is to count objects that are blended with respect to their surroundings. Due to a lack of appropriate IOC datasets, we present a large-scale dataset IOCfish5K which contains a total of 5,637 high-resolution images and 659,024 annotated center points. Our dataset consists of a large number of indiscernible objects (mainly fish) in underwater scenes, making the annotation process all the more challenging. IOCfish5K is superior to existing datasets with indiscernible scenes because of its larger scale, higher image resolutions, more annotations, and denser scenes. All these aspects make it the most challenging dataset for IOC so far, supporting progress in this area. Benefiting from the recent advancements of depth estimation foundation models, we construct high-quality depth maps for IOCfish5K by generating pseudo labels using the Depth Anything V2 model. The RGB-D version of IOCfish5K is named IOCfish5K-D. For benchmarking purposes on IOCfish5K, we select 14 mainstream methods for object counting and carefully evaluate them. For multimodal IOCfish5K-D, we evaluate other 4 popular multimodal counting methods. Furthermore, we propose IOCFormer, a new strong baseline that combines density and regression branches in a unified framework and can effectively tackle object counting under concealed scenes. We also propose IOCFormer-D to enable the effective usage of depth modality in helping detect and count objects hidden in their environments. Experiments show that IOCFormer and IOCFormer-D achieve state-of-the-art scores on IOCfish5K and IOCfish5K-D, respectively.
comment: Journal version. The resources are available at https://github.com/GuoleiSun/Indiscernible-Object-Counting
♻ ☆ CMAR-Net: Accurate Cross-Modal 3D SAR Reconstruction of Vehicle Targets with Sparse Multi-Baseline Data
Multi-baseline Synthetic Aperture Radar (SAR) three-dimensional (3D) tomography is a crucial remote sensing technique that provides 3D resolution unavailable in conventional SAR imaging. However, achieving high-quality imaging typically requires multi-angle or full-aperture data, resulting in significant imaging costs. Recent advancements in sparse 3D SAR, which rely on data from limited apertures, have gained attention as a cost-effective alternative. Notably, deep learning techniques have markedly enhanced the imaging quality of sparse 3D SAR. Despite these advancements, existing methods primarily depend on high-resolution radar images for supervising the training of deep neural networks (DNNs). This exclusive dependence on single-modal data prevents the introduction of complementary information from other data sources, limiting further improvements in imaging performance. In this paper, we introduce a Cross-Modal 3D-SAR Reconstruction Network (CMAR-Net) to enhance 3D SAR imaging by integrating heterogeneous information. Leveraging cross-modal supervision from 2D optical images and error transfer guaranteed by differentiable rendering, CMAR-Net achieves efficient training and reconstructs highly sparse multi-baseline SAR data into visually structured and accurate 3D images, particularly for vehicle targets. Extensive experiments on simulated and real-world datasets demonstrate that CMAR-Net significantly outperforms SOTA sparse reconstruction algorithms based on compressed sensing (CS) and deep learning (DL). Furthermore, our method eliminates the need for time-consuming full-aperture data preprocessing and relies solely on computer-rendered optical images, significantly reducing dataset construction costs. This work highlights the potential of deep learning for multi-baseline SAR 3D imaging and introduces a novel framework for radar imaging research through cross-modal learning.
♻ ☆ Arc2Avatar: Generating Expressive 3D Avatars from a Single Image via ID Guidance
Inspired by the effectiveness of 3D Gaussian Splatting (3DGS) in reconstructing detailed 3D scenes within multi-view setups and the emergence of large 2D human foundation models, we introduce Arc2Avatar, the first SDS-based method utilizing a human face foundation model as guidance with just a single image as input. To achieve that, we extend such a model for diverse-view human head generation by fine-tuning on synthetic data and modifying its conditioning. Our avatars maintain a dense correspondence with a human face mesh template, allowing blendshape-based expression generation. This is achieved through a modified 3DGS approach, connectivity regularizers, and a strategic initialization tailored for our task. Additionally, we propose an optional efficient SDS-based correction step to refine the blendshape expressions, enhancing realism and diversity. Experiments demonstrate that Arc2Avatar achieves state-of-the-art realism and identity preservation, effectively addressing color issues by allowing the use of very low guidance, enabled by our strong identity prior and initialization strategy, without compromising detail. Please visit https://arc2avatar.github.io for more resources.
comment: Project Page https://arc2avatar.github.io
♻ ☆ RAD-DINO: Exploring Scalable Medical Image Encoders Beyond Text Supervision
Language-supervised pre-training has proven to be a valuable method for extracting semantically meaningful features from images, serving as a foundational element in multimodal systems within the computer vision and medical imaging domains. However, the computed features are limited by the information contained in the text, which is particularly problematic in medical imaging, where the findings described by radiologists focus on specific observations. This challenge is compounded by the scarcity of paired imaging-text data due to concerns over leakage of personal health information. In this work, we fundamentally challenge the prevailing reliance on language supervision for learning general-purpose biomedical imaging encoders. We introduce RAD-DINO, a biomedical image encoder pre-trained solely on unimodal biomedical imaging data that obtains similar or greater performance than state-of-the-art biomedical language-supervised models on a diverse range of benchmarks. Specifically, the quality of learned representations is evaluated on standard imaging tasks (classification and semantic segmentation), and a vision-language alignment task (text report generation from images). To further demonstrate the drawback of language supervision, we show that features from RAD-DINO correlate with other medical records (e.g., sex or age) better than language-supervised models, which are generally not mentioned in radiology reports. Finally, we conduct a series of ablations determining the factors in RAD-DINO's performance; notably, we observe that RAD-DINO's downstream performance scales well with the quantity and diversity of training data, demonstrating that image-only supervision is a scalable approach for training a foundational biomedical image encoder. Model weights of RAD-DINO trained on publicly available datasets are available at https://huggingface.co/microsoft/rad-dino.
♻ ☆ Agentic Copyright Watermarking against Adversarial Evidence Forgery with Purification-Agnostic Curriculum Proxy Learning
With the proliferation of AI agents in various domains, protecting the ownership of AI models has become crucial due to the significant investment in their development. Unauthorized use and illegal distribution of these models pose serious threats to intellectual property, necessitating effective copyright protection measures. Model watermarking has emerged as a key technique to address this issue, embedding ownership information within models to assert rightful ownership during copyright disputes. This paper presents several contributions to model watermarking: a self-authenticating black-box watermarking protocol using hash techniques, a study on evidence forgery attacks using adversarial perturbations, a proposed defense involving a purification step to counter adversarial attacks, and a purification-agnostic curriculum proxy learning method to enhance watermark robustness and model performance. Experimental results demonstrate the effectiveness of these approaches in improving the security, reliability, and performance of watermarked models.
♻ ☆ Rethinking Decoders for Transformer-based Semantic Segmentation: A Compression Perspective NeurIPS2024
State-of-the-art methods for Transformer-based semantic segmentation typically adopt Transformer decoders that are used to extract additional embeddings from image embeddings via cross-attention, refine either or both types of embeddings via self-attention, and project image embeddings onto the additional embeddings via dot-product. Despite their remarkable success, these empirical designs still lack theoretical justifications or interpretations, thus hindering potentially principled improvements. In this paper, we argue that there are fundamental connections between semantic segmentation and compression, especially between the Transformer decoders and Principal Component Analysis (PCA). From such a perspective, we derive a white-box, fully attentional DEcoder for PrIncipled semantiC segemenTation (DEPICT), with the interpretations as follows: 1) the self-attention operator refines image embeddings to construct an ideal principal subspace that aligns with the supervision and retains most information; 2) the cross-attention operator seeks to find a low-rank approximation of the refined image embeddings, which is expected to be a set of orthonormal bases of the principal subspace and corresponds to the predefined classes; 3) the dot-product operation yields compact representation for image embeddings as segmentation masks. Experiments conducted on dataset ADE20K find that DEPICT consistently outperforms its black-box counterpart, Segmenter, and it is light weight and more robust.
comment: NeurIPS2024. Code:https://github.com/QishuaiWen/DEPICT/
♻ ☆ ScVLM: Enhancing Vision-Language Model for Safety-Critical Event Understanding WACV
Accurately identifying, understanding and describing traffic safety-critical events (SCEs), including crashes, tire strikes, and near-crashes, is crucial for advanced driver assistance systems, automated driving systems, and traffic safety. As SCEs are rare events, most general vision-language models (VLMs) have not been trained sufficiently to link SCE videos and narratives, which could lead to hallucinations and missing key safety characteristics. Here, we introduce ScVLM, a novel hybrid methodology that integrates supervised and contrastive learning techniques to classify the severity and types of SCEs, as well as to generate narrative descriptions of SCEs. This approach utilizes classification to enhance VLMs' comprehension of driving videos and improve the rationality of event descriptions. The proposed approach is trained on and evaluated by more than 8,600 SCEs from the Second Strategic Highway Research Program Naturalistic Driving Study dataset, the largest publicly accessible driving dataset with videos and SCE annotations. The results demonstrate the superiority of the proposed approach in generating contextually accurate event descriptions and mitigating VLM hallucinations. The code will be available at https://github.com/datadrivenwheels/ScVLM.
comment: To appear in Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2025
♻ ☆ Automation of Quantum Dot Measurement Analysis via Explainable Machine Learning AAAI 2024
The rapid development of quantum dot (QD) devices for quantum computing has necessitated more efficient and automated methods for device characterization and tuning. This work demonstrates the feasibility and advantages of applying explainable machine learning techniques to the analysis of quantum dot measurements, paving the way for further advances in automated and transparent QD device tuning. Many of the measurements acquired during the tuning process come in the form of images that need to be properly analyzed to guide the subsequent tuning steps. By design, features present in such images capture certain behaviors or states of the measured QD devices. When considered carefully, such features can aid the control and calibration of QD devices. An important example of such images are so-called $\textit{triangle plots}$, which visually represent current flow and reveal characteristics important for QD device calibration. While image-based classification tools, such as convolutional neural networks (CNNs), can be used to verify whether a given measurement is $\textit{good}$ and thus warrants the initiation of the next phase of tuning, they do not provide any insights into how the device should be adjusted in the case of $\textit{bad}$ images. This is because CNNs sacrifice prediction and model intelligibility for high accuracy. To ameliorate this trade-off, a recent study introduced an image vectorization approach that relies on the Gabor wavelet transform (Schug $\textit{et al.}$ 2024 $\textit{Proc. XAI4Sci: Explainable Machine Learning for Sciences Workshop (AAAI 2024) (Vancouver, Canada)}$ pp 1-6). Here we propose an alternative vectorization method that involves mathematical modeling of synthetic triangles to mimic the experimental data. Using explainable boosting machines, we show that this new method offers superior explainability of model prediction without sacrificing accuracy.
comment: 20 pages, 5 figures, abbreviated version published in Proceedings of the XAI4Sci: Explainable machine learning for sciences workshop at AAAI 2024, (Vancouver, Canada)
♻ ☆ Class Distance Weighted Cross Entropy Loss for Classification of Disease Severity
Assessing disease severity with ordinal classes, where each class reflects increasing severity levels, benefits from loss functions designed for this ordinal structure. Traditional categorical loss functions, like Cross-Entropy (CE), often perform suboptimally in these scenarios. To address this, we propose a novel loss function, Class Distance Weighted Cross-Entropy (CDW-CE), which penalizes misclassifications more severely when the predicted and actual classes are farther apart. We evaluated CDW-CE using various deep architectures, comparing its performance against several categorical and ordinal loss functions. To assess the quality of latent representations, we used t-distributed stochastic neighbor embedding (t-SNE) and uniform manifold approximation and projection (UMAP) visualizations, quantified the clustering quality using the Silhouette Score, and compared Class Activation Maps (CAM) generated by models trained with CDW-CE and CE loss. Feedback from domain experts was incorporated to evaluate how well model attention aligns with expert opinion. Our results show that CDW-CE consistently improves performance in ordinal image classification tasks. It achieves higher Silhouette Scores, indicating better class discrimination capability, and its CAM visualizations show a stronger focus on clinically significant regions, as validated by domain experts. Receiver operator characteristics (ROC) curves and the area under the curve (AUC) scores highlight that CDW-CE outperforms other loss functions, including prominent ordinal loss functions from the literature.
♻ ☆ FusionSORT: Fusion Methods for Online Multi-object Visual Tracking
In this work, we investigate four different fusion methods for associating detections to tracklets in multi-object visual tracking. In addition to considering strong cues such as motion and appearance information, we also consider weak cues such as height intersection-over-union (height-IoU) and tracklet confidence information in the data association using different fusion methods. These fusion methods include minimum, weighted sum based on IoU, Kalman filter (KF) gating, and hadamard product of costs due to the different cues. We conduct extensive evaluations on validation sets of MOT17, MOT20 and DanceTrack datasets, and find out that the choice of a fusion method is key for data association in multi-object visual tracking. We hope that this investigative work helps the computer vision research community to use the right fusion method for data association in multi-object visual tracking.
♻ ☆ Light Transport-aware Diffusion Posterior Sampling for Single-View Reconstruction of 3D Volumes
We introduce a single-view reconstruction technique of volumetric fields in which multiple light scattering effects are omnipresent, such as in clouds. We model the unknown distribution of volumetric fields using an unconditional diffusion model trained on a novel benchmark dataset comprising 1,000 synthetically simulated volumetric density fields. The neural diffusion model is trained on the latent codes of a novel, diffusion-friendly, monoplanar representation. The generative model is used to incorporate a tailored parametric diffusion posterior sampling technique into different reconstruction tasks. A physically-based differentiable volume renderer is employed to provide gradients with respect to light transport in the latent space. This stands in contrast to classic NeRF approaches and makes the reconstructions better aligned with observed data. Through various experiments, we demonstrate single-view reconstruction of volumetric clouds at a previously unattainable quality.
♻ ☆ Zero-Shot Pupil Segmentation with SAM 2: A Case Study of Over 14 Million Images
We explore the transformative potential of SAM 2, a vision foundation model, in advancing gaze estimation and eye tracking technologies. By significantly reducing annotation time, lowering technical barriers through its ease of deployment, and enhancing segmentation accuracy, SAM 2 addresses critical challenges faced by researchers and practitioners. Utilizing its zero-shot segmentation capabilities with minimal user input-a single click per video-we tested SAM 2 on over 14 million eye images from diverse datasets, including virtual reality setups and the world's largest unified dataset recorded using wearable eye trackers. Remarkably, in pupil segmentation tasks, SAM 2 matches the performance of domain-specific models trained solely on eye images, achieving competitive mean Intersection over Union (mIoU) scores of up to 93% without fine-tuning. Additionally, we provide our code and segmentation masks for these widely used datasets to promote further research.
comment: Virmarie Maquiling and Sean Anthony Byrne contributed equally to this paper, 8 pages, 3 figures, ETRA 2025, pre-print
♻ ☆ Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
We introduce InternVL 2.5, an advanced multimodal large language model (MLLM) series that builds upon InternVL 2.0, maintaining its core model architecture while introducing significant enhancements in training and testing strategies as well as data quality. In this work, we delve into the relationship between model scaling and performance, systematically exploring the performance trends in vision encoders, language models, dataset sizes, and test-time configurations. Through extensive evaluations on a wide range of benchmarks, including multi-discipline reasoning, document understanding, multi-image / video understanding, real-world comprehension, multimodal hallucination detection, visual grounding, multilingual capabilities, and pure language processing, InternVL 2.5 exhibits competitive performance, rivaling leading commercial models such as GPT-4o and Claude-3.5-Sonnet. Notably, our model is the first open-source MLLMs to surpass 70% on the MMMU benchmark, achieving a 3.7-point improvement through Chain-of-Thought (CoT) reasoning and showcasing strong potential for test-time scaling. We hope this model contributes to the open-source community by setting new standards for developing and applying multimodal AI systems. HuggingFace demo see https://huggingface.co/spaces/OpenGVLab/InternVL
comment: Technical Report
♻ ☆ BayesAdapter: enhanced uncertainty estimation in CLIP few-shot adaptation
The emergence of large pre-trained vision-language models (VLMs) represents a paradigm shift in machine learning, with unprecedented results in a broad span of visual recognition tasks. CLIP, one of the most popular VLMs, has exhibited remarkable zero-shot and transfer learning capabilities in classification. To transfer CLIP to downstream tasks, adapters constitute a parameter-efficient approach that avoids backpropagation through the large model (unlike related prompt learning methods). However, CLIP adapters have been developed to target discriminative performance, and the quality of their uncertainty estimates has been overlooked. In this work we show that the discriminative performance of state-of-the-art CLIP adapters does not always correlate with their uncertainty estimation capabilities, which are essential for a safe deployment in real-world scenarios. We also demonstrate that one of such adapters is obtained through MAP inference from a more general probabilistic framework. Based on this observation we introduce BayesAdapter, which leverages Bayesian inference to estimate a full probability distribution instead of a single point, better capturing the variability inherent in the parameter space. In a comprehensive empirical evaluation we show that our approach obtains high quality uncertainty estimates in the predictions, standing out in calibration and selective classification. Our code will be publicly available upon acceptance of the paper.
comment: 30 pages, 5 figures, 23 tables
♻ ☆ GIM: A Million-scale Benchmark for Generative Image Manipulation Detection and Localization
The extraordinary ability of generative models emerges as a new trend in image editing and generating realistic images, posing a serious threat to the trustworthiness of multimedia data and driving the research of image manipulation detection and location (IMDL). However, the lack of a large-scale data foundation makes the IMDL task unattainable. In this paper, we build a local manipulation data generation pipeline that integrates the powerful capabilities of SAM, LLM, and generative models. Upon this basis, we propose the GIM dataset, which has the following advantages: 1) Large scale, GIM includes over one million pairs of AI-manipulated images and real images. 2) Rich image content, GIM encompasses a broad range of image classes. 3) Diverse generative manipulation, the images are manipulated images with state-of-the-art generators and various manipulation tasks. The aforementioned advantages allow for a more comprehensive evaluation of IMDL methods, extending their applicability to diverse images. We introduce the GIM benchmark with two settings to evaluate existing IMDL methods. In addition, we propose a novel IMDL framework, termed GIMFormer, which consists of a ShadowTracer, Frequency-Spatial block (FSB), and a Multi-Window Anomalous Modeling (MWAM) module. Extensive experiments on the GIM demonstrate that GIMFormer surpasses the previous state-of-the-art approach on two different benchmarks.
comment: Code page: https://github.com/chenyirui/GIM
♻ ☆ Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloud
Recent advancements in self-supervised learning in the point cloud domain have demonstrated significant potential. However, these methods often suffer from drawbacks, including lengthy pre-training time, the necessity of reconstruction in the input space, or the necessity of additional modalities. In order to address these issues, we introduce Point-JEPA, a joint embedding predictive architecture designed specifically for point cloud data. To this end, we introduce a sequencer that orders point cloud patch embeddings to efficiently compute and utilize their proximity based on the indices during target and context selection. The sequencer also allows shared computations of the patch embeddings' proximity between context and target selection, further improving the efficiency. Experimentally, our method achieves competitive results with state-of-the-art methods while avoiding the reconstruction in the input space or additional modality.
comment: 13 pages, 4 figures
♻ ☆ SCC-YOLO: An Improved Object Detector for Assisting in Brain Tumor Diagnosis
Brain tumors can result in neurological dysfunction, alterations in cognitive and psychological states, increased intracranial pressure, and the occurrence of seizures, thereby presenting a substantial risk to human life and health. The You Only Look Once(YOLO) series models have demonstrated superior accuracy in object detection for medical imaging. In this paper, we develop a novel SCC-YOLO architecture by integrating the SCConv attention mechanism into YOLOv9. The SCConv module reconstructs an efficient convolutional module by reducing spatial and channel redundancy among features, thereby enhancing the learning of image features. We investigate the impact of intergrating different attention mechanisms with the YOLOv9 model on brain tumor image detection using both the Br35H dataset and our self-made dataset(Brain_Tumor_Dataset). Experimental results show that on the Br35H dataset, SCC-YOLO achieved a 0.3% improvement in mAp50 compared to YOLOv9, while on our self-made dataset, SCC-YOLO exhibited a 0.5% improvement over YOLOv9. SCC-YOLO has reached state-of-the-art performance in brain tumor detection. Source code is available at : https://jihulab.com/healthcare-information-studio/SCC-YOLO/-/tree/master
♻ ☆ Text-Guided Coarse-to-Fine Fusion Network for Robust Remote Sensing Visual Question Answering
Remote Sensing Visual Question Answering (RSVQA) has gained significant research interest. However, current RSVQA methods are limited by the imaging mechanisms of optical sensors, particularly under challenging conditions such as cloud-covered and low-light scenarios. Given the all-time and all-weather imaging capabilities of Synthetic Aperture Radar (SAR), it is crucial to investigate the integration of optical-SAR images to improve RSVQA performance. In this work, we propose a Text-guided Coarse-to-Fine Fusion Network (TGFNet), which leverages the semantic relationships between question text and multi-source images to guide the network toward complementary fusion at the feature level. Specifically, we develop a Text-guided Coarse-to-Fine Attention Refinement (CFAR) module to focus on key areas related to the question in complex remote sensing images. This module progressively directs attention from broad areas to finer details through key region routing, enhancing the model's ability to focus on relevant regions. Furthermore, we propose an Adaptive Multi-Expert Fusion (AMEF) module that dynamically integrates different experts, enabling the adaptive fusion of optical and SAR features. In addition, we create the first large-scale benchmark dataset for evaluating optical-SAR RSVQA methods, comprising 6,008 well-aligned optical-SAR image pairs and 1,036,694 well-labeled question-answer pairs across 16 diverse question types, including complex relational reasoning questions. Extensive experiments on the proposed dataset demonstrate that our TGFNet effectively integrates complementary information between optical and SAR images, significantly improving the model's performance in challenging scenarios. The dataset is available at: https://github.com/mmic-lcl/. Index Terms: Remote Sensing Visual Question Answering, Multi-source Data Fusion, Multimodal, Remote Sensing, OPT-SAR.
♻ ☆ AI-Driven Early Mental Health Screening: Analyzing Selfies of Pregnant Women ALT
Major Depressive Disorder and anxiety disorders affect millions globally, contributing significantly to the burden of mental health issues. Early screening is crucial for effective intervention, as timely identification of mental health issues can significantly improve treatment outcomes. Artificial intelligence (AI) can be valuable for improving the screening of mental disorders, enabling early intervention and better treatment outcomes. AI-driven screening can leverage the analysis of multiple data sources, including facial features in digital images. However, existing methods often rely on controlled environments or specialized equipment, limiting their broad applicability. This study explores the potential of AI models for ubiquitous depression-anxiety screening given face-centric selfies. The investigation focuses on high-risk pregnant patients, a population that is particularly vulnerable to mental health issues. To cope with limited training data resulting from our clinical setup, pre-trained models were utilized in two different approaches: fine-tuning convolutional neural networks (CNNs) originally designed for facial expression recognition and employing vision-language models (VLMs) for zero-shot analysis of facial expressions. Experimental results indicate that the proposed VLM-based method significantly outperforms CNNs, achieving an accuracy of 77.6%. Although there is significant room for improvement, the results suggest that VLMs can be a promising approach for mental health screening.
comment: This article has been accepted for publication in HEALTHINF25 at the 18th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2025)
♻ ☆ Improving Forward Compatibility in Class Incremental Learning by Increasing Representation Rank and Feature Richness
Class Incremental Learning (CIL) constitutes a pivotal subfield within continual learning, aimed at enabling models to progressively learn new classification tasks while retaining knowledge obtained from prior tasks. Although previous studies have predominantly focused on backward compatible approaches to mitigate catastrophic forgetting, recent investigations have introduced forward compatible methods to enhance performance on novel tasks and complement existing backward compatible methods. In this study, we introduce an effective-Rank based Feature Richness enhancement (RFR) method, designed for improving forward compatibility. Specifically, this method increases the effective rank of representations during the base session, thereby facilitating the incorporation of more informative features pertinent to unseen novel tasks. Consequently, RFR achieves dual objectives in backward and forward compatibility: minimizing feature extractor modifications and enhancing novel task performance, respectively. To validate the efficacy of our approach, we establish a theoretical connection between effective rank and the Shannon entropy of representations. Subsequently, we conduct comprehensive experiments by integrating RFR into eleven well-known CIL methods. Our results demonstrate the effectiveness of our approach in enhancing novel-task performance while mitigating catastrophic forgetting. Furthermore, our method notably improves the average incremental accuracy across all eleven cases examined.
♻ ☆ Benchmarking Counterfactual Image Generation NeurIPS 2024
Generative AI has revolutionised visual content editing, empowering users to effortlessly modify images and videos. However, not all edits are equal. To perform realistic edits in domains such as natural image or medical imaging, modifications must respect causal relationships inherent to the data generation process. Such image editing falls into the counterfactual image generation regime. Evaluating counterfactual image generation is substantially complex: not only it lacks observable ground truths, but also requires adherence to causal constraints. Although several counterfactual image generation methods and evaluation metrics exist, a comprehensive comparison within a unified setting is lacking. We present a comparison framework to thoroughly benchmark counterfactual image generation methods. We integrate all models that have been used for the task at hand and expand them to novel datasets and causal graphs, demonstrating the superiority of Hierarchical VAEs across most datasets and metrics. Our framework is implemented in a user-friendly Python package that can be extended to incorporate additional SCMs, causal methods, generative models, and datasets for the community to build on. Code: https://github.com/gulnazaki/counterfactual-benchmark.
comment: Published as a conference paper at NeurIPS 2024 Datasets and Benchmarks Track https://openreview.net/forum?id=0T8xRFrScB Project page: https://gulnazaki.github.io/counterfactual-benchmark
♻ ☆ Situational Scene Graph for Structured Human-centric Situation Understanding WACV 2025
Graph based representation has been widely used in modelling spatio-temporal relationships in video understanding. Although effective, existing graph-based approaches focus on capturing the human-object relationships while ignoring fine-grained semantic properties of the action components. These semantic properties are crucial for understanding the current situation, such as where does the action takes place, what tools are used and functional properties of the objects. In this work, we propose a graph-based representation called Situational Scene Graph (SSG) to encode both human-object relationships and the corresponding semantic properties. The semantic details are represented as predefined roles and values inspired by situation frame, which is originally designed to represent a single action. Based on our proposed representation, we introduce the task of situational scene graph generation and propose a multi-stage pipeline Interactive and Complementary Network (InComNet) to address the task. Given that the existing datasets are not applicable to the task, we further introduce a SSG dataset whose annotations consist of semantic role-value frames for human, objects and verb predicates of human-object relations. Finally, we demonstrate the effectiveness of our proposed SSG representation by testing on different downstream tasks. Experimental results show that the unified representation can not only benefit predicate classification and semantic role-value classification, but also benefit reasoning tasks on human-centric situation understanding. We will release the code and the dataset soon.
comment: Accepted for WACV 2025
♻ ☆ Multi-Head Explainer: A General Framework to Improve Explainability in CNNs and Transformers
In this study, we introduce the Multi-Head Explainer (MHEX), a versatile and modular framework that enhances both the explainability and accuracy of Convolutional Neural Networks (CNNs) and Transformer-based models. MHEX consists of three core components: an Attention Gate that dynamically highlights task-relevant features, Deep Supervision that guides early layers to capture fine-grained details pertinent to the target class, and an Equivalent Matrix that unifies refined local and global representations to generate comprehensive saliency maps. Our approach demonstrates superior compatibility, enabling effortless integration into existing residual networks like ResNet and Transformer architectures such as BERT with minimal modifications. Extensive experiments on benchmark datasets in medical imaging and text classification show that MHEX not only improves classification accuracy but also produces highly interpretable and detailed saliency scores.
♻ ☆ OCTolyzer: Fully automatic toolkit for segmentation and feature extracting in optical coherence tomography and scanning laser ophthalmoscopy data
Optical coherence tomography (OCT) and scanning laser ophthalmoscopy (SLO) of the eye has become essential to ophthalmology and the emerging field of oculomics, thus requiring a need for transparent, reproducible, and rapid analysis of this data for clinical research and the wider research community. Here, we introduce OCTolyzer, the first open-source toolkit for retinochoroidal analysis in OCT/SLO data. It features two analysis suites for OCT and SLO data, facilitating deep learning-based anatomical segmentation and feature extraction of the cross-sectional retinal and choroidal layers and en face retinal vessels. We describe OCTolyzer and evaluate the reproducibility of its OCT choroid analysis. At the population level, metrics for choroid region thickness were highly reproducible, with a mean absolute error (MAE)/Pearson correlation for macular volume choroid thickness (CT) of 6.7$\mu$m/0.99, macular B-scan CT of 11.6$\mu$m/0.99, and peripapillary CT of 5.0$\mu$m/0.99. Macular choroid vascular index (CVI) also showed strong reproducibility, with MAE/Pearson for volume CVI yielding 0.0271/0.97 and B-scan CVI 0.0130/0.91. At the eye level, measurement noise for regional and vessel metrics was below 5% and 20% of the population's variability, respectively. Outliers were caused by poor-quality B-scans with thick choroids and invisible choroid-sclera boundary. Processing times on a laptop CPU were under three seconds for macular/peripapillary B-scans and 85 seconds for volume scans. OCTolyzer can convert OCT/SLO data into reproducible and clinically meaningful retinochoroidal features and will improve the standardisation of ocular measurements in OCT/SLO image analysis, requiring no specialised training or proprietary software to be used. OCTolyzer is freely available here: https://github.com/jaburke166/OCTolyzer.
comment: Main paper: 15 pages, 9 figures, 3 tables. Supplementary material: 9 pages, 6 figures, 5 tables
♻ ☆ VibrantVS: A high-resolution multi-task transformer for forest canopy height estimation
This paper explores the application of a novel multi-task vision transformer (ViT) model for the estimation of canopy height models (CHMs) using 4-band National Agriculture Imagery Program (NAIP) imagery across the western United States. We compare the effectiveness of this model in terms of accuracy and precision aggregated across ecoregions and class heights versus three other benchmark peer-reviewed models. Key findings suggest that, while other benchmark models can provide high precision in localized areas, the VibrantVS model has substantial advantages across a broad reach of ecoregions in the western United States with higher accuracy, higher precision, the ability to generate updated inference at a cadence of three years or less, and high spatial resolution. The VibrantVS model provides significant value for ecological monitoring and land management decisions for wildfire mitigation.
comment: 15 pages, 12 figures
♻ ☆ SyncDiff: Synchronized Motion Diffusion for Multi-Body Human-Object Interaction Synthesis
Synthesizing realistic human-object interaction motions is a critical problem in VR/AR and human animation. Unlike the commonly studied scenarios involving a single human or hand interacting with one object, we address a more generic multi-body setting with arbitrary numbers of humans, hands, and objects. This complexity introduces significant challenges in synchronizing motions due to the high correlations and mutual influences among bodies. To address these challenges, we introduce SyncDiff, a novel method for multi-body interaction synthesis using a synchronized motion diffusion strategy. SyncDiff employs a single diffusion model to capture the joint distribution of multi-body motions. To enhance motion fidelity, we propose a frequency-domain motion decomposition scheme. Additionally, we introduce a new set of alignment scores to emphasize the synchronization of different body motions. SyncDiff jointly optimizes both data sample likelihood and alignment likelihood through an explicit synchronization strategy. Extensive experiments across four datasets with various multi-body configurations demonstrate the superiority of SyncDiff over existing state-of-the-art motion synthesis methods.
♻ ☆ PSA-VLM: Enhancing Vision-Language Model Safety through Progressive Concept-Bottleneck-Driven Alignment
Benefiting from the powerful capabilities of Large Language Models (LLMs), pre-trained visual encoder models connected to LLMs form Vision Language Models (VLMs). However, recent research shows that the visual modality in VLMs is highly vulnerable, allowing attackers to bypass safety alignment in LLMs through visually transmitted content, launching harmful attacks. To address this challenge, we propose a progressive concept-based alignment strategy, PSA-VLM, which incorporates safety modules as concept bottlenecks to enhance visual modality safety alignment. By aligning model predictions with specific safety concepts, we improve defenses against risky images, enhancing explainability and controllability while minimally impacting general performance. Our method is obtained through two-stage training. The low computational cost of the first stage brings very effective performance improvement, and the fine-tuning of the language model in the second stage further improves the safety performance. Our method achieves state-of-the-art results on popular VLM safety benchmark.
comment: arXiv admin note: substantial text overlap with arXiv:2405.13581
♻ ☆ Migician: Revealing the Magic of Free-Form Multi-Image Grounding in Multimodal Large Language Models
The recent advancement of Multimodal Large Language Models (MLLMs) has significantly improved their fine-grained perception of single images and general comprehension across multiple images. However, existing MLLMs still face challenges in achieving precise grounding in complex multi-image scenarios. To address this, we first explore a Chain-of-Thought (CoT) framework that integrates single-image grounding with multi-image comprehension. While partially effective, it remains unstable and struggles to capture abstract visual information due to its non-end-to-end nature. Therefore, we introduce Migician, the first multi-image grounding model capable of performing free-form and accurate grounding across multiple images. To support this, we present the MGrounding-630k dataset, which comprises data for several multi-image grounding tasks derived from existing datasets, along with newly generated free-form grounding instruction-following data. Furthermore, we propose MIG-Bench, a comprehensive benchmark specifically designed for evaluating multi-image grounding capabilities. Experimental results demonstrate that our model achieves significantly superior multi-image grounding capabilities, outperforming the best existing MLLMs by 21.61% and even surpassing much larger 70B models. Our code, model, dataset, and benchmark are fully open-sourced at https://migician-vg.github.io/.
comment: 20 pages, 8 figures
Amortizing intractable inference in diffusion models for vision, language, and control NeurIPS 2024
Diffusion models have emerged as effective distribution estimators in vision, language, and reinforcement learning, but their use as priors in downstream tasks poses an intractable posterior inference problem. This paper studies amortized sampling of the posterior over data, $\mathbf{x}\sim p^{\rm post}(\mathbf{x})\propto p(\mathbf{x})r(\mathbf{x})$, in a model that consists of a diffusion generative model prior $p(\mathbf{x})$ and a black-box constraint or likelihood function $r(\mathbf{x})$. We state and prove the asymptotic correctness of a data-free learning objective, relative trajectory balance, for training a diffusion model that samples from this posterior, a problem that existing methods solve only approximately or in restricted cases. Relative trajectory balance arises from the generative flow network perspective on diffusion models, which allows the use of deep reinforcement learning techniques to improve mode coverage. Experiments illustrate the broad potential of unbiased inference of arbitrary posteriors under diffusion priors: in vision (classifier guidance), language (infilling under a discrete diffusion LLM), and multimodal data (text-to-image generation). Beyond generative modeling, we apply relative trajectory balance to the problem of continuous control with a score-based behavior prior, achieving state-of-the-art results on benchmarks in offline reinforcement learning.
comment: NeurIPS 2024; code: https://github.com/GFNOrg/diffusion-finetuning
♻ ☆ InstructOCR: Instruction Boosting Scene Text Spotting AAAI2025
In the field of scene text spotting, previous OCR methods primarily relied on image encoders and pre-trained text information, but they often overlooked the advantages of incorporating human language instructions. To address this gap, we propose InstructOCR, an innovative instruction-based scene text spotting model that leverages human language instructions to enhance the understanding of text within images. Our framework employs both text and image encoders during training and inference, along with instructions meticulously designed based on text attributes. This approach enables the model to interpret text more accurately and flexibly. Extensive experiments demonstrate the effectiveness of our model and we achieve state-of-the-art results on widely used benchmarks. Furthermore, the proposed framework can be seamlessly applied to scene text VQA tasks. By leveraging instruction strategies during pre-training, the performance on downstream VQA tasks can be significantly improved, with a 2.6% increase on the TextVQA dataset and a 2.1% increase on the ST-VQA dataset. These experimental results provide insights into the benefits of incorporating human language instructions for OCR-related tasks.
comment: Accepted by AAAI2025
♻ ☆ II-Bench: An Image Implication Understanding Benchmark for Multimodal Large Language Models
The rapid advancements in the development of multimodal large language models (MLLMs) have consistently led to new breakthroughs on various benchmarks. In response, numerous challenging and comprehensive benchmarks have been proposed to more accurately assess the capabilities of MLLMs. However, there is a dearth of exploration of the higher-order perceptual capabilities of MLLMs. To fill this gap, we propose the Image Implication understanding Benchmark, II-Bench, which aims to evaluate the model's higher-order perception of images. Through extensive experiments on II-Bench across multiple MLLMs, we have made significant findings. Initially, a substantial gap is observed between the performance of MLLMs and humans on II-Bench. The pinnacle accuracy of MLLMs attains 74.8%, whereas human accuracy averages 90%, peaking at an impressive 98%. Subsequently, MLLMs perform worse on abstract and complex images, suggesting limitations in their ability to understand high-level semantics and capture image details. Finally, it is observed that most models exhibit enhanced accuracy when image sentiment polarity hints are incorporated into the prompts. This observation underscores a notable deficiency in their inherent understanding of image sentiment. We believe that II-Bench will inspire the community to develop the next generation of MLLMs, advancing the journey towards expert artificial general intelligence (AGI). II-Bench is publicly available at https://huggingface.co/datasets/m-a-p/II-Bench.
comment: 100 pages, 82 figures, add citations
♻ ☆ EM-DARTS: Hierarchical Differentiable Architecture Search for Eye Movement Recognition
Eye movement biometrics has received increasing attention thanks to its highly secure identification. Although deep learning (DL) models have shown success in eye movement recognition, their architectures largely rely on human prior knowledge. Differentiable Neural Architecture Search (DARTS) automates the manual process of architecture design with high search efficiency. However, DARTS typically stacks multiple cells to form a convolutional network, which limits the diversity of architecture. Furthermore, DARTS generally searches for architectures using shallower networks than those used in the evaluation, creating a significant disparity in architecture depth between the search and evaluation phases. To address this issue, we propose EM-DARTS, a hierarchical differentiable architecture search algorithm to automatically design the DL architecture for eye movement recognition. First, we define a supernet and propose a global and local alternate Neural Architecture Search method to search the optimal architecture alternately with a differentiable neural architecture search. The local search strategy aims to find an optimal architecture for different cells while the global search strategy is responsible for optimizing the architecture of the target network. To minimize redundancy, transfer entropy is proposed to compute the information amount of each layer, thereby further simplifying the network search process. Experimental results on three public datasets demonstrate that the proposed EM-DARTS is capable of producing an optimal architecture that leads to state-of-the-art recognition performance, {Specifically, the recognition models developed using EM-DARTS achieved the lowest EERs of 0.0453 on the GazeBase dataset, 0.0377 on the JuDo1000 dataset, and 0.1385 on the EMglasses dataset.
comment: Submited to IEEE Transactions on Instrumentation and Measurement
♻ ☆ WeCromCL: Weakly Supervised Cross-Modality Contrastive Learning for Transcription-only Supervised Text Spotting ECCV 2024
Transcription-only Supervised Text Spotting aims to learn text spotters relying only on transcriptions but no text boundaries for supervision, thus eliminating expensive boundary annotation. The crux of this task lies in locating each transcription in scene text images without location annotations. In this work, we formulate this challenging problem as a Weakly Supervised Cross-modality Contrastive Learning problem, and design a simple yet effective model dubbed WeCromCL that is able to detect each transcription in a scene image in a weakly supervised manner. Unlike typical methods for cross-modality contrastive learning that focus on modeling the holistic semantic correlation between an entire image and a text description, our WeCromCL conducts atomistic contrastive learning to model the character-wise appearance consistency between a text transcription and its correlated region in a scene image to detect an anchor point for the transcription in a weakly supervised manner. The detected anchor points by WeCromCL are further used as pseudo location labels to guide the learning of text spotting. Extensive experiments on four challenging benchmarks demonstrate the superior performance of our model over other methods. Code will be released.
comment: Accepted by ECCV 2024
♻ ☆ AI-Driven Diabetic Retinopathy Screening: Multicentric Validation of AIDRSS in India
Purpose: Diabetic retinopathy (DR) is a major cause of vision loss, particularly in India, where access to retina specialists is limited in rural areas. This study aims to evaluate the Artificial Intelligence-based Diabetic Retinopathy Screening System (AIDRSS) for DR detection and prevalence assessment, addressing the growing need for scalable, automated screening solutions in resource-limited settings. Approach: A multicentric, cross-sectional study was conducted in Kolkata, India, involving 5,029 participants and 10,058 macula-centric retinal fundus images. The AIDRSS employed a deep learning algorithm with 50 million trainable parameters, integrated with Contrast Limited Adaptive Histogram Equalization (CLAHE) preprocessing for enhanced image quality. DR was graded using the International Clinical Diabetic Retinopathy (ICDR) Scale, categorizing disease into five stages (DR0 to DR4). Statistical metrics including sensitivity, specificity, and prevalence rates were evaluated against expert retina specialist assessments. Results: The prevalence of DR in the general population was 13.7%, rising to 38.2% among individuals with elevated random blood glucose levels. The AIDRSS achieved an overall sensitivity of 92%, specificity of 88%, and 100% sensitivity for detecting referable DR (DR3 and DR4). These results demonstrate the system's robust performance in accurately identifying and grading DR in a diverse population. Conclusions: AIDRSS provides a reliable, scalable solution for early DR detection in resource-constrained environments. Its integration of advanced AI techniques ensures high diagnostic accuracy, with potential to significantly reduce the burden of diabetes-related vision loss in underserved regions.
comment: 22 pages, 5 figures. arXiv admin note: substantial text overlap with arXiv:1812.07105 by other authors without attribution
♻ ☆ HeadGAP: Few-Shot 3D Head Avatar via Generalizable Gaussian Priors 3DV 2025
In this paper, we present a novel 3D head avatar creation approach capable of generalizing from few-shot in-the-wild data with high-fidelity and animatable robustness. Given the underconstrained nature of this problem, incorporating prior knowledge is essential. Therefore, we propose a framework comprising prior learning and avatar creation phases. The prior learning phase leverages 3D head priors derived from a large-scale multi-view dynamic dataset, and the avatar creation phase applies these priors for few-shot personalization. Our approach effectively captures these priors by utilizing a Gaussian Splatting-based auto-decoder network with part-based dynamic modeling. Our method employs identity-shared encoding with personalized latent codes for individual identities to learn the attributes of Gaussian primitives. During the avatar creation phase, we achieve fast head avatar personalization by leveraging inversion and fine-tuning strategies. Extensive experiments demonstrate that our model effectively exploits head priors and successfully generalizes them to few-shot personalization, achieving photo-realistic rendering quality, multi-view consistency, and stable animation.
comment: Accepted to 3DV 2025. Project page: https://headgap.github.io/
♻ ☆ Quilt-LLaVA: Visual Instruction Tuning by Extracting Localized Narratives from Open-Source Histopathology Videos
Diagnosis in histopathology requires a global whole slide images (WSIs) analysis, requiring pathologists to compound evidence from different WSI patches. The gigapixel scale of WSIs poses a challenge for histopathology multi-modal models. Training multi-model models for histopathology requires instruction tuning datasets, which currently contain information for individual image patches, without a spatial grounding of the concepts within each patch and without a wider view of the WSI. Therefore, they lack sufficient diagnostic capacity for histopathology. To bridge this gap, we introduce Quilt-Instruct, a large-scale dataset of 107,131 histopathology-specific instruction question/answer pairs, grounded within diagnostically relevant image patches that make up the WSI. Our dataset is collected by leveraging educational histopathology videos from YouTube, which provides spatial localization of narrations by automatically extracting the narrators' cursor positions. Quilt-Instruct supports contextual reasoning by extracting diagnosis and supporting facts from the entire WSI. Using Quilt-Instruct, we train Quilt-LLaVA, which can reason beyond the given single image patch, enabling diagnostic reasoning across patches. To evaluate Quilt-LLaVA, we propose a comprehensive evaluation dataset created from 985 images and 1283 human-generated question-answers. We also thoroughly evaluate Quilt-LLaVA using public histopathology datasets, where Quilt-LLaVA significantly outperforms SOTA by over 10% on relative GPT-4 score and 4% and 9% on open and closed set VQA. Our code, data, and model are publicly accessible at quilt-llava.github.io.
♻ ☆ Simplifying CLIP: Unleashing the Power of Large-Scale Models on Consumer-level Computers
Contrastive Language-Image Pre-training (CLIP) has attracted a surge of attention for its superior zero-shot performance and excellent transferability to downstream tasks. However, training such large-scale models usually requires substantial computation and storage, which poses barriers for general users with consumer-level computers. Motivated by this observation, in this paper we investigate how to achieve competitive performance on only one Nvidia RTX3090 GPU and with one terabyte for storing dataset. On one hand, we simplify the transformer block structure and combine Weight Inheritance with multi-stage Knowledge Distillation (WIKD), thereby reducing the parameters and improving the inference speed during training along with deployment. On the other hand, confronted with the convergence challenge posed by small dataset, we generate synthetic captions for each sample as data augmentation, and devise a novel Pair Matching (PM) loss to fully exploit the distinguishment among positive and negative image-text pairs. Extensive experiments demonstrate that our model can achieve a new state-of-the-art datascale-parameter-accuracy tradeoff, which could further popularize the CLIP model in the related research community.
♻ ☆ Buster: Implanting Semantic Backdoor into Text Encoder to Mitigate NSFW Content Generation
The rise of deep learning models in the digital era has raised substantial concerns regarding the generation of Not-Safe-for-Work (NSFW) content. Existing defense methods primarily involve model fine-tuning and post-hoc content moderation. Nevertheless, these approaches largely lack scalability in eliminating harmful content, degrade the quality of benign image generation, or incur high inference costs. To address these challenges, we propose an innovative framework named \textit{Buster}, which injects backdoors into the text encoder to prevent NSFW content generation. Buster leverages deep semantic information rather than explicit prompts as triggers, redirecting NSFW prompts towards targeted benign prompts. Additionally, Buster employs energy-based training data generation through Langevin dynamics for adversarial knowledge augmentation, thereby ensuring robustness in harmful concept definition. This approach demonstrates exceptional resilience and scalability in mitigating NSFW content. Particularly, Buster fine-tunes the text encoder of Text-to-Image models within merely five minutes, showcasing its efficiency. Our extensive experiments denote that Buster outperforms nine state-of-the-art baselines, achieving a superior NSFW content removal rate of at least 91.2\% while preserving the quality of harmless images.
♻ ☆ On the Robustness of Object Detection Models on Aerial Images
The robustness of object detection models is a major concern when applied to real-world scenarios. The performance of most models tends to degrade when confronted with images affected by corruptions, since they are usually trained and evaluated on clean datasets. While numerous studies have explored the robustness of object detection models on natural images, there is a paucity of research focused on models applied to aerial images, which feature complex backgrounds, substantial variations in scales, and orientations of objects. This paper addresses the challenge of assessing the robustness of object detection models on aerial images, with a specific emphasis on scenarios where images are affected by clouds. In this study, we introduce two novel benchmarks based on DOTA-v1.0. The first benchmark encompasses 19 prevalent corruptions, while the second focuses on the cloud-corrupted condition-a phenomenon uncommon in natural images yet frequent in aerial photography. We systematically evaluate the robustness of mainstream object detection models and perform necessary ablation experiments. Through our investigations, we find that rotation-invariant modeling and enhanced backbone architectures can improve the robustness of models. Furthermore, increasing the capacity of Transformer-based backbones can strengthen their robustness. The benchmarks we propose and our comprehensive experimental analyses can facilitate research on robust object detection on aerial images. The codes and datasets are available at: https://github.com/hehaodong530/DOTA-C.
comment: accepted by IEEE TGRS
♻ ☆ Pamba: Enhancing Global Interaction in Point Clouds via State Space Model AAAI 2025
Transformers have demonstrated impressive results for 3D point cloud semantic segmentation. However, the quadratic complexity of transformer makes computation costs high, limiting the number of points that can be processed simultaneously and impeding the modeling of long-range dependencies between objects in a single scene. Drawing inspiration from the great potential of recent state space models (SSM) for long sequence modeling, we introduce Mamba, an SSM-based architecture, to the point cloud domain and propose Pamba, a novel architecture with strong global modeling capability under linear complexity. Specifically, to make the disorderness of point clouds fit in with the causal nature of Mamba, we propose a multi-path serialization strategy applicable to point clouds. Besides, we propose the ConvMamba block to compensate for the shortcomings of Mamba in modeling local geometries and in unidirectional modeling. Pamba obtains state-of-the-art results on several 3D point cloud segmentation tasks, including ScanNet v2, ScanNet200, S3DIS and nuScenes, while its effectiveness is validated by extensive experiments.
comment: Accepted by AAAI 2025
♻ ☆ MovieCharacter: A Tuning-Free Framework for Controllable Character Video Synthesis
Recent advancements in character video synthesis still depend on extensive fine-tuning or complex 3D modeling processes, which can restrict accessibility and hinder real-time applicability. To address these challenges, we propose a simple yet effective tuning-free framework for character video synthesis, named MovieCharacter, designed to streamline the synthesis process while ensuring high-quality outcomes. Our framework decomposes the synthesis task into distinct, manageable modules: character segmentation and tracking, video object removal, character motion imitation, and video composition. This modular design not only facilitates flexible customization but also ensures that each component operates collaboratively to effectively meet user needs. By leveraging existing open-source models and integrating well-established techniques, MovieCharacter achieves impressive synthesis results without necessitating substantial resources or proprietary datasets. Experimental results demonstrate that our framework enhances the efficiency, accessibility, and adaptability of character video synthesis, paving the way for broader creative and interactive applications.
♻ ☆ MLLM-CompBench: A Comparative Reasoning Benchmark for Multimodal LLMs NeurIPS 2024
The ability to compare objects, scenes, or situations is crucial for effective decision-making and problem-solving in everyday life. For instance, comparing the freshness of apples enables better choices during grocery shopping while comparing sofa designs helps optimize the aesthetics of our living space. Despite its significance, the comparative capability is largely unexplored in artificial general intelligence (AGI). In this paper, we introduce MLLM-CompBench, a benchmark designed to evaluate the comparative reasoning capability of multimodal large language models (MLLMs). MLLM-CompBench mines and pairs images through visually oriented questions covering eight dimensions of relative comparison: visual attribute, existence, state, emotion, temporality, spatiality, quantity, and quality. We curate a collection of around 40K image pairs using metadata from diverse vision datasets and CLIP similarity scores. These image pairs span a broad array of visual domains, including animals, fashion, sports, and both outdoor and indoor scenes. The questions are carefully crafted to discern relative characteristics between two images and are labeled by human annotators for accuracy and relevance. We use MLLM-CompBench to evaluate recent MLLMs, including GPT-4V(ision), Gemini-Pro, and LLaVA-1.6. Our results reveal notable shortcomings in their comparative abilities. We believe MLLM-COMPBENCH not only sheds light on these limitations but also establishes a solid foundation for future enhancements in the comparative capability of MLLMs.
comment: This paper has been accepted to NeurIPS 2024. The first two authors contributed equally to this work
♻ ☆ SL-YOLO: A Stronger and Lighter Drone Target Detection Model
Detecting small objects in complex scenes, such as those captured by drones, is a daunting challenge due to the difficulty in capturing the complex features of small targets. While the YOLO family has achieved great success in large target detection, its performance is less than satisfactory when faced with small targets. Because of this, this paper proposes a revolutionary model SL-YOLO (Stronger and Lighter YOLO) that aims to break the bottleneck of small target detection. We propose the Hierarchical Extended Path Aggregation Network (HEPAN), a pioneering cross-scale feature fusion method that can ensure unparalleled detection accuracy even in the most challenging environments. At the same time, without sacrificing detection capabilities, we design the C2fDCB lightweight module and add the SCDown downsampling module to greatly reduce the model's parameters and computational complexity. Our experimental results on the VisDrone2019 dataset reveal a significant improvement in performance, with mAP@0.5 jumping from 43.0% to 46.9% and mAP@0.5:0.95 increasing from 26.0% to 28.9%. At the same time, the model parameters are reduced from 11.1M to 9.6M, and the FPS can reach 132, making it an ideal solution for real-time small object detection in resource-constrained environments.
♻ ☆ SoftPatch+: Fully Unsupervised Anomaly Classification and Segmentation
Although mainstream unsupervised anomaly detection (AD) (including image-level classification and pixel-level segmentation)algorithms perform well in academic datasets, their performance is limited in practical application due to the ideal experimental setting of clean training data. Training with noisy data is an inevitable problem in real-world anomaly detection but is seldom discussed. This paper is the first to consider fully unsupervised industrial anomaly detection (i.e., unsupervised AD with noisy data). To solve this problem, we proposed memory-based unsupervised AD methods, SoftPatch and SoftPatch+, which efficiently denoise the data at the patch level. Noise discriminators are utilized to generate outlier scores for patch-level noise elimination before coreset construction. The scores are then stored in the memory bank to soften the anomaly detection boundary. Compared with existing methods, SoftPatch maintains a strong modeling ability of normal data and alleviates the overconfidence problem in coreset, and SoftPatch+ has more robust performance which is articularly useful in real-world industrial inspection scenarios with high levels of noise (from 10% to 40%). Comprehensive experiments conducted in diverse noise scenarios demonstrate that both SoftPatch and SoftPatch+ outperform the state-of-the-art AD methods on the MVTecAD, ViSA, and BTAD benchmarks. Furthermore, the performance of SoftPatch and SoftPatch+ is comparable to that of the noise-free methods in conventional unsupervised AD setting. The code of the proposed methods can be found at https://github.com/TencentYoutuResearch/AnomalyDetection-SoftPatch.
comment: arXiv admin note: substantial text overlap with arXiv:2403.14233 paper has been accepted by Pattern Recognition
♻ ☆ MedicalNarratives: Connecting Medical Vision and Language with Localized Narratives
We propose MedicalNarratives, a dataset curated from medical pedagogical videos similar in nature to data collected in Think-Aloud studies and inspired by Localized Narratives, which collects grounded image-text data by curating instructors' speech and mouse cursor movements synchronized in time. MedicalNarratives enables pretraining of both semantic and dense objectives, alleviating the need to train medical semantic and dense tasks disparately due to the lack of reasonably sized datasets. Our dataset contains 4.7M image-text pairs from videos and articles, with 1M samples containing dense annotations in the form of traces and bounding boxes. To evaluate the utility of MedicalNarratives, we train GenMedClip based on the CLIP architecture using our dataset spanning 12 medical domains and demonstrate that it outperforms previous state-of-the-art models on a newly constructed medical imaging benchmark that comprehensively evaluates performance across all modalities. Data, demo, code and models available at https://medical-narratives.github.io
♻ ☆ Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking Multimodal Large Language Models ECCV 2024
In this paper, we study the harmlessness alignment problem of multimodal large language models (MLLMs). We conduct a systematic empirical analysis of the harmlessness performance of representative MLLMs and reveal that the image input poses the alignment vulnerability of MLLMs. Inspired by this, we propose a novel jailbreak method named HADES, which hides and amplifies the harmfulness of the malicious intent within the text input, using meticulously crafted images. Experimental results show that HADES can effectively jailbreak existing MLLMs, which achieves an average Attack Success Rate (ASR) of 90.26% for LLaVA-1.5 and 71.60% for Gemini Pro Vision. Our code and data are available at https://github.com/RUCAIBox/HADES.
comment: ECCV 2024 Oral
♻ ☆ The Streetscape Application Services Stack (SASS): Towards a Distributed Sensing Architecture for Urban Applications
As urban populations grow, cities are becoming more complex, driving the deployment of interconnected sensing systems to realize the vision of smart cities. These systems aim to improve safety, mobility, and quality of life through applications that integrate diverse sensors with real-time decision-making. Streetscape applications-focusing on challenges like pedestrian safety and adaptive traffic management-depend on managing distributed, heterogeneous sensor data, aligning information across time and space, and enabling real-time processing. These tasks are inherently complex and often difficult to scale. The Streetscape Application Services Stack (SASS) addresses these challenges with three core services: multimodal data synchronization, spatiotemporal data fusion, and distributed edge computing. By structuring these capabilities as clear, composable abstractions with clear semantics, SASS allows developers to scale streetscape applications efficiently while minimizing the complexity of multimodal integration. We evaluated SASS in two real-world testbed environments: a controlled parking lot and an urban intersection in a major U.S. city. These testbeds allowed us to test SASS under diverse conditions, demonstrating its practical applicability. The Multimodal Data Synchronization service reduced temporal misalignment errors by 88%, achieving synchronization accuracy within 50 milliseconds. Spatiotemporal Data Fusion service improved detection accuracy for pedestrians and vehicles by over 10%, leveraging multicamera integration. The Distributed Edge Computing service increased system throughput by more than an order of magnitude. Together, these results show how SASS provides the abstractions and performance needed to support real-time, scalable urban applications, bridging the gap between sensing infrastructure and actionable streetscape intelligence.
♻ ☆ Valley2: Exploring Multimodal Models with Scalable Vision-Language Design
Recently, vision-language models have made remarkable progress, demonstrating outstanding capabilities in various tasks such as image captioning and video understanding. We introduce Valley2, a novel multimodal large language model designed to enhance performance across all domains and extend the boundaries of practical applications in e-commerce and short video scenarios. Notably, Valley2 achieves state-of-the-art (SOTA) performance on e-commerce benchmarks, surpassing open-source models of similar size by a large margin (79.66 vs. 72.76). Additionally, Valley2 ranks second on the OpenCompass leaderboard among models with fewer than 10B parameters, with an impressive average score of 67.4. The code and model weights are open-sourced at https://github.com/bytedance/Valley.
♻ ☆ ChartX & ChartVLM: A Versatile Benchmark and Foundation Model for Complicated Chart Reasoning
Recently, many versatile Multi-modal Large Language Models (MLLMs) have emerged continuously. However, their capacity to query information depicted in visual charts and engage in reasoning based on the queried contents remains under-explored. In this paper, to comprehensively and rigorously benchmark the ability of the off-the-shelf MLLMs in the chart domain, we construct ChartX, a multi-modal evaluation set covering 18 chart types, 7 chart tasks, 22 disciplinary topics, and high-quality chart data. Besides, we develop ChartVLM to offer a new perspective on handling multi-modal tasks that strongly depend on interpretable patterns, such as reasoning tasks in the field of charts or geometric images. We evaluate the chart-related ability of mainstream MLLMs and our ChartVLM on the proposed ChartX evaluation set. Extensive experiments demonstrate that ChartVLM surpasses both versatile and chart-related large models, achieving results comparable to GPT-4V. We believe that our study can pave the way for further exploration in creating a more comprehensive chart evaluation set and developing more interpretable multi-modal models. Both ChartX and ChartVLM are available at: https://github.com/Alpha-Innovator/ChartVLM
comment: Code and dataset are available for downloading at: https://github.com/Alpha-Innovator/ChartVLM 25 pages, 15 figures
♻ ☆ LDMapNet-U: An End-to-End System for City-Scale Lane-Level Map Updating KDD 2025
An up-to-date city-scale lane-level map is an indispensable infrastructure and a key enabling technology for ensuring the safety and user experience of autonomous driving systems. In industrial scenarios, reliance on manual annotation for map updates creates a critical bottleneck. Lane-level updates require precise change information and must ensure consistency with adjacent data while adhering to strict standards. Traditional methods utilize a three-stage approach-construction, change detection, and updating-which often necessitates manual verification due to accuracy limitations. This results in labor-intensive processes and hampers timely updates. To address these challenges, we propose LDMapNet-U, which implements a new end-to-end paradigm for city-scale lane-level map updating. By reconceptualizing the update task as an end-to-end map generation process grounded in historical map data, we introduce a paradigm shift in map updating that simultaneously generates vectorized maps and change information. To achieve this, a Prior-Map Encoding (PME) module is introduced to effectively encode historical maps, serving as a critical reference for detecting changes. Additionally, we incorporate a novel Instance Change Prediction (ICP) module that learns to predict associations with historical maps. Consequently, LDMapNet-U simultaneously achieves vectorized map element generation and change detection. To demonstrate the superiority and effectiveness of LDMapNet-U, extensive experiments are conducted using large-scale real-world datasets. In addition, LDMapNet-U has been successfully deployed in production at Baidu Maps since April 2024, supporting map updating for over 360 cities and significantly shortening the update cycle from quarterly to weekly. The updated maps serve hundreds of millions of users and are integrated into the autonomous driving systems of several leading vehicle companies.
comment: Accepted by KDD 2025, camera-ready version
♻ ☆ PViT: Prior-augmented Vision Transformer for Out-of-distribution Detection
Vision Transformers (ViTs) have achieved remarkable success over various vision tasks, yet their robustness against data distribution shifts and inherent inductive biases remain underexplored. To enhance the robustness of ViT models for image Out-of-Distribution (OOD) detection, we introduce a novel and generic framework named Prior-augmented Vision Transformer (PViT). Taking as input the prior class logits from a pretrained model, we train PViT to predict the class logits. During inference, PViT identifies OOD samples by quantifying the divergence between the predicted class logits and the prior logits obtained from pre-trained models. Unlike existing state-of-the-art(SOTA) OOD detection methods, PViT shapes the decision boundary between ID and OOD by utilizing the proposed prior guided confidence, without requiring additional data modeling, generation methods, or structural modifications. Extensive experiments on the large-scale ImageNet benchmark, evaluated against over seven OOD datasets, demonstrate that PViT significantly outperforms existing SOTA OOD detection methods in terms of FPR95 and AUROC. The codebase is publicly available at https://github.com/RanchoGoose/PViT.
♻ ☆ Sparse Attention Vectors: Generative Multimodal Model Features Are Discriminative Vision-Language Classifiers
Generative Large Multimodal Models (LMMs) like LLaVA and Qwen-VL excel at a wide variety of vision-language (VL) tasks such as image captioning or visual question answering. Despite strong performance, LMMs are not directly suited for foundational discriminative vision-language tasks (i.e., tasks requiring discrete label predictions) such as image classification and multiple-choice VQA. One key challenge in utilizing LMMs for discriminative tasks is the extraction of useful features from generative models. To overcome this issue, we propose an approach for finding features in the model's latent space to more effectively leverage LMMs for discriminative tasks. Toward this end, we present Sparse Attention Vectors (SAVs) -- a finetuning-free method that leverages sparse attention head activations (fewer than 1\% of the heads) in LMMs as strong features for VL tasks. With only few-shot examples, SAVs demonstrate state-of-the-art performance compared to a variety of few-shot and finetuned baselines on a collection of discriminative tasks. Our experiments also imply that SAVs can scale in performance with additional examples and generalize to similar tasks, establishing SAVs as both effective and robust multimodal feature representations.
♻ ☆ Pre-trained Vision-Language Models Learn Discoverable Visual Concepts
Do vision-language models (VLMs) pre-trained to caption an image of a "durian" learn visual concepts such as "brown" (color) and "spiky" (texture) at the same time? We aim to answer this question as visual concepts learned "for free" would enable wide applications such as neuro-symbolic reasoning or human-interpretable object classification. We assume that the visual concepts, if captured by pre-trained VLMs, can be extracted by their vision-language interface with text-based concept prompts. We observe that recent works prompting VLMs with concepts often differ in their strategies to define and evaluate the visual concepts, leading to conflicting conclusions. We propose a new concept definition strategy based on two observations: First, certain concept prompts include shortcuts that recognize correct concepts for wrong reasons; Second, multimodal information (e.g. visual discriminativeness, and textual knowledge) should be leveraged when selecting the concepts. Our proposed concept discovery and learning (CDL) framework is thus designed to identify a diverse list of generic visual concepts (e.g. "spiky" as opposed to "spiky durian"), which are ranked and selected based on visual and language mutual information. We carefully design quantitative and human evaluations of the discovered concepts on six diverse visual recognition datasets, which confirm that pre-trained VLMs do learn visual concepts that provide accurate and thorough descriptions for the recognized objects. All code and models are publicly released.
comment: Transactions on Machine Learning Research, 2025
♻ ☆ Extracting Manifold Information from Point Clouds
A kernel based method is proposed for the construction of signature (defining) functions of subsets of $\mathbb{R}^d$. The subsets can range from full dimensional manifolds (open subsets) to point clouds (a finite number of points) and include bounded smooth manifolds of any codimension. The interpolation and analysis of point clouds are the main application. Two extreme cases in terms of regularity are considered, where the data set is interpolated by an analytic surface, at the one extreme, and by a H\"older continuous surface, at the other. The signature function can be computed as a linear combination of translated kernels, the coefficients of which are the solution of a finite dimensional linear problem. Once it is obtained, it can be used to estimate the dimension as well as the normal and the curvatures of the interpolated surface. The method is global and does not require explicit knowledge of local neighborhoods or any other structure present in the data set. It admits a variational formulation with a natural ``regularized'' counterpart, that proves to be useful in dealing with data sets corrupted by numerical error or noise. The underlying analytical structure of the approach is presented in general before it is applied to the case of point clouds.
comment: 27 pages, 16 figures, 5 tables
♻ ☆ ExACT: Teaching AI Agents to Explore with Reflective-MCTS and Exploratory Learning
Autonomous agents have demonstrated significant potential in automating complex multistep decision-making tasks. However, even state-of-the-art vision-language models (VLMs), such as GPT-4o, still fall short of human-level performance, particularly in intricate web environments and long-horizon tasks. To address these limitations, we present ExACT, an approach to combine test-time search and self-learning to build o1-like models for agentic applications. We first introduce Reflective Monte Carlo Tree Search (R-MCTS), a novel test time algorithm designed to enhance AI agents' ability to explore decision space on the fly. R-MCTS extends traditional MCTS by 1) incorporating contrastive reflection, allowing agents to learn from past interactions and dynamically improve their search efficiency; and 2) using multi-agent debate for reliable state evaluation. Next, we introduce Exploratory Learning, a novel learning strategy to teach agents to search at inference time without relying on any external search algorithms. On the challenging VisualWebArena benchmark, our GPT-4o based R-MCTS agent achieves a 6% to 30% relative improvement across various tasks compared to the previous state-of-the-art. Additionally, we show that the knowledge and experience gained from test-time search can be effectively transferred back to GPT-4o via fine-tuning. After Exploratory Learning, GPT-4o 1) demonstrates the ability to explore the environment, evaluate a state, and backtrack to viable ones when it detects that the current state cannot lead to success, and 2) matches 87% of R-MCTS's performance while using significantly less compute. Notably, our work demonstrates the compute scaling properties in both training - data collection with R-MCTS - and testing time. These results suggest a promising research direction to enhance VLMs' capabilities for agentic applications via test-time search and self-learning.
Computational Geometry 7
☆ Computational Geometry with Probabilistically Noisy Primitive Operations
Much prior work has been done on designing computational geometry algorithms that handle input degeneracies, data imprecision, and arithmetic round-off errors. We take a new approach, inspired by the noisy sorting literature, and study computational geometry algorithms subject to noisy Boolean primitive operations in which, e.g., the comparison "is point q above line L?" returns the wrong answer with some fixed probability. We propose a novel technique called path-guided pushdown random walks that generalizes the results of noisy sorting. We apply this technique to solve point-location, plane-sweep, convex hulls in 2D and 3D, dynamic 2D convex hulls, and Delaunay triangulations for noisy primitives in optimal time with high probability.
comment: 28 pages, 6 figures
☆ A Greedy Algorithm for Low-Crossing Partitions for General Set Systems
Simplicial partitions are a fundamental structure in computational geometry, as they form the basis of optimal data structures for range searching and several related problems. Current algorithms are built on very specific spatial partitioning tools tailored for certain geometric cases. This severely limits their applicability to general set systems. In this work, we propose a simple greedy heuristic for constructing simplicial partitions of any set system. We present a thorough empirical evaluation of its behavior on a variety of geometric and non-geometric set systems, showing that it performs well on most instances. Implementation of these algorithms is available on Github.
♻ ☆ A cohomology-based Gromov-Hausdorff metric approach for quantifying molecular similarity
We introduce, for the first time, a cohomology-based Gromov-Hausdorff ultrametric method to analyze 1-dimensional and higher-dimensional (co)homology groups, focusing on loops, voids, and higher-dimensional cavity structures in simplicial complexes, to address typical clustering questions arising in molecular data analysis. The Gromov-Hausdorff distance quantifies the dissimilarity between two metric spaces. In this framework, molecules are represented as simplicial complexes, and their cohomology vector spaces are computed to capture intrinsic topological invariants encoding loop and cavity structures. These vector spaces are equipped with a suitable distance measure, enabling the computation of the Gromov-Hausdorff ultrametric to evaluate structural dissimilarities. We demonstrate the methodology using organic-inorganic halide perovskite (OIHP) structures. The results highlight the effectiveness of this approach in clustering various molecular structures. By incorporating geometric information, our method provides deeper insights compared to traditional persistent homology techniques.
comment: 14 pages, 3 figures
♻ ☆ Ortho-unit polygons can be guarded with at most $\lfloor \frac{n-4}{8} \rfloor$ guards
An orthogonal polygon is called an ortho-unit polygon if its vertices have integer coordinates, and all of its edges have length one. In this paper we prove that any ortho-unit polygon with $n \geq 12$ vertices can be guarded with at most $\lfloor \frac{n-4}{8} \rfloor$ guards, which is a tight bound.
comment: 15 pages, 11 figures
♻ ☆ Extracting Manifold Information from Point Clouds
A kernel based method is proposed for the construction of signature (defining) functions of subsets of $\mathbb{R}^d$. The subsets can range from full dimensional manifolds (open subsets) to point clouds (a finite number of points) and include bounded smooth manifolds of any codimension. The interpolation and analysis of point clouds are the main application. Two extreme cases in terms of regularity are considered, where the data set is interpolated by an analytic surface, at the one extreme, and by a H\"older continuous surface, at the other. The signature function can be computed as a linear combination of translated kernels, the coefficients of which are the solution of a finite dimensional linear problem. Once it is obtained, it can be used to estimate the dimension as well as the normal and the curvatures of the interpolated surface. The method is global and does not require explicit knowledge of local neighborhoods or any other structure present in the data set. It admits a variational formulation with a natural ``regularized'' counterpart, that proves to be useful in dealing with data sets corrupted by numerical error or noise. The underlying analytical structure of the approach is presented in general before it is applied to the case of point clouds.
comment: 27 pages, 16 figures, 5 tables
♻ ☆ Construction, Transformation and Structures of 2x2 Space-Filling Curves
The 2x2 space-filling curve is a type of generalized space-filling curve characterized by a basic unit is in a "U-shape" that traverses a 2x2 grid. In this work, we propose a universal framework for constructing general 2x2 curves where self-similarity is not strictly required. The construction is based on a novel set of grammars that define the expansion of curves from level 0 (a single point) to level 1 (units in U-shapes), which ultimately determines all $36 \times 2^k$ possible forms of curves on any level $k$ initialized from single points. We further developed an encoding system in which each unique form of the curve is associated with a specific combination of an initial seed and a sequence of codes that sufficiently describes both the global and local structures of the curve. We demonstrated that this encoding system is a powerful tool for studying 2x2 curves and we established comprehensive theoretical foundations from the following three key perspectives: 1) We provided a deterministic encoding for any unit on any level and position on the curve, enabling the study of curve generation across arbitrary parts on the curve and ranges of iterations; 2) We gave deterministic encodings for various curve transformations, including rotations, reflections and reversals; 3) We provided deterministic forms of families of curves exhibiting specific structures, including homogeneous curves, curves with identical shapes, and with completely distinct shapes. We also explored families of recursive curves, subunit identically shaped curves, symmetric curves and closed curves. Finally, we proposed a method to calculate the location of any point on the curve arithmetically, within a time complexity linear to the level of the curve.
♻ ☆ HyperSteiner: Computing Heuristic Hyperbolic Steiner Minimal Trees
We propose HyperSteiner -- an efficient heuristic algorithm for computing Steiner minimal trees in the hyperbolic space. HyperSteiner extends the Euclidean Smith-Lee-Liebman algorithm, which is grounded in a divide-and-conquer approach involving the Delaunay triangulation. The central idea is rephrasing Steiner tree problems with three terminals as a system of equations in the Klein-Beltrami model. Motivated by the fact that hyperbolic geometry is well-suited for representing hierarchies, we explore applications to hierarchy discovery in data. Results show that HyperSteiner infers more realistic hierarchies than the Minimum Spanning Tree and is more scalable to large datasets than Neighbor Joining.
Computation and Language 52
☆ WebWalker: Benchmarking LLMs in Web Traversal
Retrieval-augmented generation (RAG) demonstrates remarkable performance across tasks in open-domain question-answering. However, traditional search engines may retrieve shallow content, limiting the ability of LLMs to handle complex, multi-layered information. To address it, we introduce WebWalkerQA, a benchmark designed to assess the ability of LLMs to perform web traversal. It evaluates the capacity of LLMs to traverse a website's subpages to extract high-quality data systematically. We propose WebWalker, which is a multi-agent framework that mimics human-like web navigation through an explore-critic paradigm. Extensive experimental results show that WebWalkerQA is challenging and demonstrates the effectiveness of RAG combined with WebWalker, through the horizontal and vertical integration in real-world scenarios.
☆ SST-EM: Advanced Metrics for Evaluating Semantic, Spatial and Temporal Aspects in Video Editing WACV
Video editing models have advanced significantly, but evaluating their performance remains challenging. Traditional metrics, such as CLIP text and image scores, often fall short: text scores are limited by inadequate training data and hierarchical dependencies, while image scores fail to assess temporal consistency. We present SST-EM (Semantic, Spatial, and Temporal Evaluation Metric), a novel evaluation framework that leverages modern Vision-Language Models (VLMs), Object Detection, and Temporal Consistency checks. SST-EM comprises four components: (1) semantic extraction from frames using a VLM, (2) primary object tracking with Object Detection, (3) focused object refinement via an LLM agent, and (4) temporal consistency assessment using a Vision Transformer (ViT). These components are integrated into a unified metric with weights derived from human evaluations and regression analysis. The name SST-EM reflects its focus on Semantic, Spatial, and Temporal aspects of video evaluation. SST-EM provides a comprehensive evaluation of semantic fidelity and temporal smoothness in video editing. The source code is available in the \textbf{\href{https://github.com/custommetrics-sst/SST_CustomEvaluationMetrics.git}{GitHub Repository}}.
comment: WACV workshop
☆ Imagine while Reasoning in Space: Multimodal Visualization-of-Thought
Chain-of-Thought (CoT) prompting has proven highly effective for enhancing complex reasoning in Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs). Yet, it struggles in complex spatial reasoning tasks. Nonetheless, human cognition extends beyond language alone, enabling the remarkable capability to think in both words and images. Inspired by this mechanism, we propose a new reasoning paradigm, Multimodal Visualization-of-Thought (MVoT). It enables visual thinking in MLLMs by generating image visualizations of their reasoning traces. To ensure high-quality visualization, we introduce token discrepancy loss into autoregressive MLLMs. This innovation significantly improves both visual coherence and fidelity. We validate this approach through several dynamic spatial reasoning tasks. Experimental results reveal that MVoT demonstrates competitive performance across tasks. Moreover, it exhibits robust and reliable improvements in the most challenging scenarios where CoT fails. Ultimately, MVoT establishes new possibilities for complex reasoning tasks where visual thinking can effectively complement verbal reasoning.
comment: 11 pages, 6 figures, 4 tables (27 pages, 10 figures, 16 tables including references and appendices)
☆ Investigating Large Language Models in Inferring Personality Traits from User Conversations
Large Language Models (LLMs) are demonstrating remarkable human like capabilities across diverse domains, including psychological assessment. This study evaluates whether LLMs, specifically GPT-4o and GPT-4o mini, can infer Big Five personality traits and generate Big Five Inventory-10 (BFI-10) item scores from user conversations under zero-shot prompting conditions. Our findings reveal that incorporating an intermediate step--prompting for BFI-10 item scores before calculating traits--enhances accuracy and aligns more closely with the gold standard than direct trait inference. This structured approach underscores the importance of leveraging psychological frameworks in improving predictive precision. Additionally, a group comparison based on depressive symptom presence revealed differential model performance. Participants were categorized into two groups: those experiencing at least one depressive symptom and those without symptoms. GPT-4o mini demonstrated heightened sensitivity to depression-related shifts in traits such as Neuroticism and Conscientiousness within the symptom-present group, whereas GPT-4o exhibited strengths in nuanced interpretation across groups. These findings underscore the potential of LLMs to analyze real-world psychological data effectively, offering a valuable foundation for interdisciplinary research at the intersection of artificial intelligence and psychology.
comment: 13 pages, 5 figures
☆ Parallel Key-Value Cache Fusion for Position Invariant RAG
Recent advancements in Large Language Models (LLMs) underscore the necessity of Retrieval Augmented Generation (RAG) to leverage external information. However, LLMs are sensitive to the position of relevant information within contexts and tend to generate incorrect responses when such information is placed in the middle, known as `Lost in the Middle' phenomenon. In this paper, we introduce a framework that generates consistent outputs for decoder-only models, irrespective of the input context order. Experimental results for three open domain question answering tasks demonstrate position invariance, where the model is not sensitive to input context order, and superior robustness to irrelevent passages compared to prevailing approaches for RAG pipelines.
comment: 5 pages
☆ TiEBe: A Benchmark for Assessing the Current Knowledge of Large Language Models
In a rapidly evolving knowledge landscape and the increasing adoption of large language models, a need has emerged to keep these models continuously updated with current events. While existing benchmarks evaluate general factual recall, they often overlook two critical aspects: the ability of models to integrate evolving knowledge through continual learning and the significant regional disparities in their performance. To address these gaps, we introduce the Timely Events Benchmark (TiEBe), a dataset containing over 11,000 question-answer pairs focused on globally and regionally significant events. TiEBe leverages structured retrospective data from Wikipedia, enabling continuous updates to assess LLMs' knowledge of evolving global affairs and their understanding of events across different regions. Our benchmark demonstrates that LLMs exhibit substantial geographic disparities in factual recall, emphasizing the need for more balanced global knowledge representation. Furthermore, TiEBe serves as a tool for evaluating continual learning strategies, providing insights into models' ability to acquire new information without forgetting past knowledge.
☆ Enhancing Retrieval-Augmented Generation: A Study of Best Practices
Retrieval-Augmented Generation (RAG) systems have recently shown remarkable advancements by integrating retrieval mechanisms into language models, enhancing their ability to produce more accurate and contextually relevant responses. However, the influence of various components and configurations within RAG systems remains underexplored. A comprehensive understanding of these elements is essential for tailoring RAG systems to complex retrieval tasks and ensuring optimal performance across diverse applications. In this paper, we develop several advanced RAG system designs that incorporate query expansion, various novel retrieval strategies, and a novel Contrastive In-Context Learning RAG. Our study systematically investigates key factors, including language model size, prompt design, document chunk size, knowledge base size, retrieval stride, query expansion techniques, Contrastive In-Context Learning knowledge bases, multilingual knowledge bases, and Focus Mode retrieving relevant context at sentence-level. Through extensive experimentation, we provide a detailed analysis of how these factors influence response quality. Our findings offer actionable insights for developing RAG systems, striking a balance between contextual richness and retrieval-generation efficiency, thereby paving the way for more adaptable and high-performing RAG frameworks in diverse real-world scenarios. Our code and implementation details are publicly available.
☆ Emergent effects of scaling on the functional hierarchies within large language models
Large language model (LLM) architectures are often described as functionally hierarchical: Early layers process syntax, middle layers begin to parse semantics, and late layers integrate information. The present work revisits these ideas. This research submits simple texts to an LLM (e.g., "A church and organ") and extracts the resulting activations. Then, for each layer, support vector machines and ridge regressions are fit to predict a text's label and thus examine whether a given layer encodes some information. Analyses using a small model (Llama-3.2-3b; 28 layers) partly bolster the common hierarchical perspective: Item-level semantics are most strongly represented early (layers 2-7), then two-item relations (layers 8-12), and then four-item analogies (layers 10-15). Afterward, the representation of items and simple relations gradually decreases in deeper layers that focus on more global information. However, several findings run counter to a steady hierarchy view: First, although deep layers can represent document-wide abstractions, deep layers also compress information from early portions of the context window without meaningful abstraction. Second, when examining a larger model (Llama-3.3-70b-Instruct), stark fluctuations in abstraction level appear: As depth increases, two-item relations and four-item analogies initially increase in their representation, then markedly decrease, and afterward increase again momentarily. This peculiar pattern consistently emerges across several experiments. Third, another emergent effect of scaling is coordination between the attention mechanisms of adjacent layers. Across multiple experiments using the larger model, adjacent layers fluctuate between what information they each specialize in representing. In sum, an abstraction hierarchy often manifests across layers, but large models also deviate from this structure in curious ways.
☆ Joint Automatic Speech Recognition And Structure Learning For Better Speech Understanding ICASSP 2025
Spoken language understanding (SLU) is a structure prediction task in the field of speech. Recently, many works on SLU that treat it as a sequence-to-sequence task have achieved great success. However, This method is not suitable for simultaneous speech recognition and understanding. In this paper, we propose a joint speech recognition and structure learning framework (JSRSL), an end-to-end SLU model based on span, which can accurately transcribe speech and extract structured content simultaneously. We conduct experiments on name entity recognition and intent classification using the Chinese dataset AISHELL-NER and the English dataset SLURP. The results show that our proposed method not only outperforms the traditional sequence-to-sequence method in both transcription and extraction capabilities but also achieves state-of-the-art performance on the two datasets.
comment: 5 pages, 2 figures, accepted by ICASSP 2025
☆ FinerWeb-10BT: Refining Web Data with LLM-Based Line-Level Filtering
Data quality is crucial for training Large Language Models (LLMs). Traditional heuristic filters often miss low-quality text or mistakenly remove valuable content. In this paper, we introduce an LLM-based line-level filtering method to enhance training data quality. We use GPT-4o mini to label a 20,000-document sample from FineWeb at the line level, allowing the model to create descriptive labels for low-quality lines. These labels are grouped into nine main categories, and we train a DeBERTa-v3 classifier to scale the filtering to a 10B-token subset of FineWeb. To test the impact of our filtering, we train GPT-2 models on both the original and the filtered datasets. The results show that models trained on the filtered data achieve higher accuracy on the HellaSwag benchmark and reach their performance targets faster, even with up to 25\% less data. This demonstrates that LLM-based line-level filtering can significantly improve data quality and training efficiency for LLMs. We release our quality-annotated dataset, FinerWeb-10BT, and the codebase to support further work in this area.
comment: 11 pages, 4 figures, 4 tables. To be published in NoDaLiDa/Baltic-HLT 2025 proceedings
☆ The Lessons of Developing Process Reward Models in Mathematical Reasoning
Process Reward Models (PRMs) emerge as a promising approach for process supervision in mathematical reasoning of Large Language Models (LLMs), which aim to identify and mitigate intermediate errors in the reasoning processes. However, the development of effective PRMs faces significant challenges, particularly in data annotation and evaluation methodologies. In this paper, through extensive experiments, we demonstrate that commonly used Monte Carlo (MC) estimation-based data synthesis for PRMs typically yields inferior performance and generalization compared to LLM-as-a-judge and human annotation methods. MC estimation relies on completion models to evaluate current-step correctness, leading to inaccurate step verification. Furthermore, we identify potential biases in conventional Best-of-N (BoN) evaluation strategies for PRMs: (1) The unreliable policy models generate responses with correct answers but flawed processes, leading to a misalignment between the evaluation criteria of BoN and the PRM objectives of process verification. (2) The tolerance of PRMs of such responses leads to inflated BoN scores. (3) Existing PRMs have a significant proportion of minimum scores concentrated on the final answer steps, revealing the shift from process to outcome-based assessment in BoN Optimized PRMs. To address these challenges, we develop a consensus filtering mechanism that effectively integrates MC estimation with LLM-as-a-judge and advocates a more comprehensive evaluation framework that combines response-level and step-level metrics. Based on the mechanisms, we significantly improve both model performance and data efficiency in the BoN evaluation and the step-wise error identification task. Finally, we release a new state-of-the-art PRM that outperforms existing open-source alternatives and provides practical guidelines for future research in building process supervision models.
☆ Comparative analysis of optical character recognition methods for Sámi texts from the National Library of Norway
Optical Character Recognition (OCR) is crucial to the National Library of Norway's (NLN) digitisation process as it converts scanned documents into machine-readable text. However, for the S\'ami documents in NLN's collection, the OCR accuracy is insufficient. Given that OCR quality affects downstream processes, evaluating and improving OCR for text written in S\'ami languages is necessary to make these resources accessible. To address this need, this work fine-tunes and evaluates three established OCR approaches, Transkribus, Tesseract and TrOCR, for transcribing S\'ami texts from NLN's collection. Our results show that Transkribus and TrOCR outperform Tesseract on this task, while Tesseract achieves superior performance on an out-of-domain dataset. Furthermore, we show that fine-tuning pre-trained models and supplementing manual annotations with machine annotations and synthetic text images can yield accurate OCR for S\'ami languages, even with a moderate amount of manually annotated data.
comment: To be published in Proceedings of the 25th Nordic Conference on Computational Linguistics (NoDaLiDa)
☆ Audio-CoT: Exploring Chain-of-Thought Reasoning in Large Audio Language Model
Large Audio-Language Models (LALMs) have demonstrated remarkable performance in tasks involving audio perception and understanding, such as speech recognition and audio captioning. However, their reasoning capabilities - critical for solving complex real-world problems - remain underexplored. In this work, we conduct the first exploration into integrating Chain-of-Thought (CoT) reasoning into LALMs to enhance their reasoning ability across auditory modalities. We evaluate representative CoT methods, analyzing their performance in both information extraction and reasoning tasks across sound, music, and speech domains. Our findings reveal that CoT methods significantly improve performance on easy and medium tasks but encounter challenges with hard tasks, where reasoning chains can confuse the model rather than improve accuracy. Additionally, we identify a positive correlation between reasoning path length and accuracy, demonstrating the potential of scaling inference for advanced instruction-following and reasoning. This study not only highlights the promise of CoT in enhancing LALM reasoning capabilities but also identifies key limitations and provides actionable directions for future research.
☆ Can Vision-Language Models Evaluate Handwritten Math?
Recent advancements in Vision-Language Models (VLMs) have opened new possibilities in automatic grading of handwritten student responses, particularly in mathematics. However, a comprehensive study to test the ability of VLMs to evaluate and reason over handwritten content remains absent. To address this gap, we introduce FERMAT, a benchmark designed to assess the ability of VLMs to detect, localize and correct errors in handwritten mathematical content. FERMAT spans four key error dimensions - computational, conceptual, notational, and presentation - and comprises over 2,200 handwritten math solutions derived from 609 manually curated problems from grades 7-12 with intentionally introduced perturbations. Using FERMAT we benchmark nine VLMs across three tasks: error detection, localization, and correction. Our results reveal significant shortcomings in current VLMs in reasoning over handwritten text, with Gemini-1.5-Pro achieving the highest error correction rate (77%). We also observed that some models struggle with processing handwritten content, as their accuracy improves when handwritten inputs are replaced with printed text or images. These findings highlight the limitations of current VLMs and reveal new avenues for improvement. We release FERMAT and all the associated resources in the open-source to drive further research.
☆ When lies are mostly truthful: automated verbal deception detection for embedded lies
Background: Verbal deception detection research relies on narratives and commonly assumes statements as truthful or deceptive. A more realistic perspective acknowledges that the veracity of statements exists on a continuum with truthful and deceptive parts being embedded within the same statement. However, research on embedded lies has been lagging behind. Methods: We collected a novel dataset of 2,088 truthful and deceptive statements with annotated embedded lies. Using a within-subjects design, participants provided a truthful account of an autobiographical event. They then rewrote their statement in a deceptive manner by including embedded lies, which they highlighted afterwards and judged on lie centrality, deceptiveness, and source. Results: We show that a fined-tuned language model (Llama-3-8B) can classify truthful statements and those containing embedded lies with 64% accuracy. Individual differences, linguistic properties and explainability analysis suggest that the challenge of moving the dial towards embedded lies stems from their resemblance to truthful statements. Typical deceptive statements consisted of 2/3 truthful information and 1/3 embedded lies, largely derived from past personal experiences and with minimal linguistic differences with their truthful counterparts. Conclusion: We present this dataset as a novel resource to address this challenge and foster research on embedded lies in verbal deception detection.
☆ BIOMEDICA: An Open Biomedical Image-Caption Archive, Dataset, and Vision-Language Models Derived from Scientific Literature
The development of vision-language models (VLMs) is driven by large-scale and diverse multimodal datasets. However, progress toward generalist biomedical VLMs is limited by the lack of annotated, publicly accessible datasets across biology and medicine. Existing efforts are restricted to narrow domains, missing the full diversity of biomedical knowledge encoded in scientific literature. To address this gap, we introduce BIOMEDICA, a scalable, open-source framework to extract, annotate, and serialize the entirety of the PubMed Central Open Access subset into an easy-to-use, publicly accessible dataset.Our framework produces a comprehensive archive with over 24 million unique image-text pairs from over 6 million articles. Metadata and expert-guided annotations are also provided. We demonstrate the utility and accessibility of our resource by releasing BMCA-CLIP, a suite of CLIP-style models continuously pre-trained on the BIOMEDICA dataset via streaming, eliminating the need to download 27 TB of data locally.On average, our models achieve state-of-the-art performance across 40 tasks - spanning pathology, radiology, ophthalmology, dermatology, surgery, molecular biology, parasitology, and cell biology - excelling in zero-shot classification with a 6.56% average improvement (as high as 29.8% and 17.5% in dermatology and ophthalmology, respectively), and stronger image-text retrieval, all while using 10x less compute. To foster reproducibility and collaboration, we release our codebase and dataset for the broader research community.
☆ ListConRanker: A Contrastive Text Reranker with Listwise Encoding
Reranker models aim to re-rank the passages based on the semantics similarity between the given query and passages, which have recently received more attention due to the wide application of the Retrieval-Augmented Generation. Most previous methods apply pointwise encoding, meaning that it can only encode the context of the query for each passage input into the model. However, for the reranker model, given a query, the comparison results between passages are even more important, which is called listwise encoding. Besides, previous models are trained using the cross-entropy loss function, which leads to issues of unsmooth gradient changes during training and low training efficiency. To address these issues, we propose a novel Listwise-encoded Contrastive text reRanker (ListConRanker). It can help the passage to be compared with other passages during the encoding process, and enhance the contrastive information between positive examples and between positive and negative examples. At the same time, we use the circle loss to train the model to increase the flexibility of gradients and solve the problem of training efficiency. Experimental results show that ListConRanker achieves state-of-the-art performance on the reranking benchmark of Chinese Massive Text Embedding Benchmark, including the cMedQA1.0, cMedQA2.0, MMarcoReranking, and T2Reranking datasets.
comment: 11 pages, 4 figures
☆ AdaCS: Adaptive Normalization for Enhanced Code-Switching ASR ICASSP 2025
Intra-sentential code-switching (CS) refers to the alternation between languages that happens within a single utterance and is a significant challenge for Automatic Speech Recognition (ASR) systems. For example, when a Vietnamese speaker uses foreign proper names or specialized terms within their speech. ASR systems often struggle to accurately transcribe intra-sentential CS due to their training on monolingual data and the unpredictable nature of CS. This issue is even more pronounced for low-resource languages, where limited data availability hinders the development of robust models. In this study, we propose AdaCS, a normalization model integrates an adaptive bias attention module (BAM) into encoder-decoder network. This novel approach provides a robust solution to CS ASR in unseen domains, thereby significantly enhancing our contribution to the field. By utilizing BAM to both identify and normalize CS phrases, AdaCS enhances its adaptive capabilities with a biased list of words provided during inference. Our method demonstrates impressive performance and the ability to handle unseen CS phrases across various domains. Experiments show that AdaCS outperforms previous state-of-the-art method on Vietnamese CS ASR normalization by considerable WER reduction of 56.2% and 36.8% on the two proposed test sets.
comment: Accepted at ICASSP 2025
☆ Boosting Text-To-Image Generation via Multilingual Prompting in Large Multimodal Models ICASSP 2025
Previous work on augmenting large multimodal models (LMMs) for text-to-image (T2I) generation has focused on enriching the input space of in-context learning (ICL). This includes providing a few demonstrations and optimizing image descriptions to be more detailed and logical. However, as demand for more complex and flexible image descriptions grows, enhancing comprehension of input text within the ICL paradigm remains a critical yet underexplored area. In this work, we extend this line of research by constructing parallel multilingual prompts aimed at harnessing the multilingual capabilities of LMMs. More specifically, we translate the input text into several languages and provide the models with both the original text and the translations. Experiments on two LMMs across 3 benchmarks show that our method, PMT2I, achieves superior performance in general, compositional, and fine-grained assessments, especially in human preference alignment. Additionally, with its advantage of generating more diverse images, PMT2I significantly outperforms baseline prompts when incorporated with reranking methods. Our code and parallel multilingual data can be found at https://github.com/takagi97/PMT2I.
comment: Accepted to ICASSP 2025
☆ Research on the Online Update Method for Retrieval-Augmented Generation (RAG) Model with Incremental Learning
In the contemporary context of rapid advancements in information technology and the exponential growth of data volume, language models are confronted with significant challenges in effectively navigating the dynamic and ever-evolving information landscape to update and adapt to novel knowledge in real time. In this work, an online update method is proposed, which is based on the existing Retrieval Enhanced Generation (RAG) model with multiple innovation mechanisms. Firstly, the dynamic memory is used to capture the emerging data samples, and then gradually integrate them into the core model through a tunable knowledge distillation strategy. At the same time, hierarchical indexing and multi-layer gating mechanism are introduced into the retrieval module to ensure that the retrieved content is more targeted and accurate. Finally, a multi-stage network structure is established for different types of inputs in the generation stage, and cross-attention matching and screening are carried out on the intermediate representations of each stage to ensure the effective integration and iterative update of new and old knowledge. Experimental results show that the proposed method is better than the existing mainstream comparison models in terms of knowledge retention and inference accuracy.
☆ Leveraging ASIC AI Chips for Homomorphic Encryption
Cloud-based services are making the outsourcing of sensitive client data increasingly common. Although homomorphic encryption (HE) offers strong privacy guarantee, it requires substantially more resources than computing on plaintext, often leading to unacceptably large latencies in getting the results. HE accelerators have emerged to mitigate this latency issue, but with the high cost of ASICs. In this paper we show that HE primitives can be converted to AI operators and accelerated on existing ASIC AI accelerators, like TPUs, which are already widely deployed in the cloud. Adapting such accelerators for HE requires (1) supporting modular multiplication, (2) high-precision arithmetic in software, and (3) efficient mapping on matrix engines. We introduce the CROSS compiler (1) to adopt Barrett reduction to provide modular reduction support using multiplier and adder, (2) Basis Aligned Transformation (BAT) to convert high-precision multiplication as low-precision matrix-vector multiplication, (3) Matrix Aligned Transformation (MAT) to covert vectorized modular operation with reduction into matrix multiplication that can be efficiently processed on 2D spatial matrix engine. Our evaluation of CROSS on a Google TPUv4 demonstrates significant performance improvements, with up to 161x and 5x speedup compared to the previous work on many-core CPUs and V100. The kernel-level codes are open-sourced at https://github.com/google/jaxite.git.
comment: 16 pages, 10 figures, 4 algorithms, 7 tables. Enabling Google TPUv4 for privacy-preserving AI inference
☆ ViSoLex: An Open-Source Repository for Vietnamese Social Media Lexical Normalization COLING 2025
ViSoLex is an open-source system designed to address the unique challenges of lexical normalization for Vietnamese social media text. The platform provides two core services: Non-Standard Word (NSW) Lookup and Lexical Normalization, enabling users to retrieve standard forms of informal language and standardize text containing NSWs. ViSoLex's architecture integrates pre-trained language models and weakly supervised learning techniques to ensure accurate and efficient normalization, overcoming the scarcity of labeled data in Vietnamese. This paper details the system's design, functionality, and its applications for researchers and non-technical users. Additionally, ViSoLex offers a flexible, customizable framework that can be adapted to various datasets and research requirements. By publishing the source code, ViSoLex aims to contribute to the development of more robust Vietnamese natural language processing tools and encourage further research in lexical normalization. Future directions include expanding the system's capabilities for additional languages and improving the handling of more complex non-standard linguistic patterns.
comment: The 31st International Conference on Computational Linguistics (COLING 2025)
☆ LEO: Boosting Mixture of Vision Encoders for Multimodal Large Language Models
Enhanced visual understanding serves as a cornerstone for multimodal large language models (MLLMs). Recent hybrid MLLMs incorporate a mixture of vision experts to address the limitations of using a single vision encoder and excessively long visual tokens. Despite the progress of these MLLMs, a research gap remains in effectively integrating diverse vision encoders. This work explores fusion strategies of visual tokens for hybrid MLLMs, leading to the design of LEO, a novel MLLM with a dual-branch vision encoder framework that incorporates a post-adaptation fusion strategy and adaptive tiling: for each segmented tile of the input images, LEO sequentially interleaves the visual tokens from its two vision encoders. Extensive evaluation across 13 vision-language benchmarks reveals that LEO outperforms state-of-the-art open-source MLLMs and hybrid MLLMs on the majority of tasks. Furthermore, we show that LEO can be adapted to the specialized domain of autonomous driving without altering the model architecture or training recipe, achieving competitive performance compared to existing baselines. The code and model will be publicly available.
☆ A Heterogeneous Multimodal Graph Learning Framework for Recognizing User Emotions in Social Networks
The rapid expansion of social media platforms has provided unprecedented access to massive amounts of multimodal user-generated content. Comprehending user emotions can provide valuable insights for improving communication and understanding of human behaviors. Despite significant advancements in Affective Computing, the diverse factors influencing user emotions in social networks remain relatively understudied. Moreover, there is a notable lack of deep learning-based methods for predicting user emotions in social networks, which could be addressed by leveraging the extensive multimodal data available. This work presents a novel formulation of personalized emotion prediction in social networks based on heterogeneous graph learning. Building upon this formulation, we design HMG-Emo, a Heterogeneous Multimodal Graph Learning Framework that utilizes deep learning-based features for user emotion recognition. Additionally, we include a dynamic context fusion module in HMG-Emo that is capable of adaptively integrating the different modalities in social media data. Through extensive experiments, we demonstrate the effectiveness of HMG-Emo and verify the superiority of adopting a graph neural network-based approach, which outperforms existing baselines that use rich hand-crafted features. To the best of our knowledge, HMG-Emo is the first multimodal and deep-learning-based approach to predict personalized emotions within online social networks. Our work highlights the significance of exploiting advanced deep learning techniques for less-explored problems in Affective Computing.
☆ Advancing Student Writing Through Automated Syntax Feedback
This study underscores the pivotal role of syntax feedback in augmenting the syntactic proficiency of students. Recognizing the challenges faced by learners in mastering syntactic nuances, we introduce a specialized dataset named Essay-Syntax-Instruct designed to enhance the understanding and application of English syntax among these students. Leveraging the capabilities of Large Language Models (LLMs) such as GPT3.5-Turbo, Llama-2-7b-chat-hf, Llama-2-13b-chat-hf, and Mistral-7B-Instruct-v0.2, this work embarks on a comprehensive fine-tuning process tailored to the syntax improvement task. Through meticulous evaluation, we demonstrate that the fine-tuned LLMs exhibit a marked improvement in addressing syntax-related challenges, thereby serving as a potent tool for students to identify and rectify their syntactic errors. The findings not only highlight the effectiveness of the proposed dataset in elevating the performance of LLMs for syntax enhancement but also illuminate a promising path for utilizing advanced language models to support language acquisition efforts. This research contributes to the broader field of language learning technology by showcasing the potential of LLMs in facilitating the linguistic development of Students.
comment: This paper has been accepted for presentation at AIEER 2024
☆ Exploring the encoding of linguistic representations in the Fully-Connected Layer of generative CNNs for Speech
Interpretability work on the convolutional layers of CNNs has primarily focused on computer vision, but some studies also explore correspondences between the latent space and the output in the audio domain. However, it has not been thoroughly examined how acoustic and linguistic information is represented in the fully connected (FC) layer that bridges the latent space and convolutional layers. The current study presents the first exploration of how the FC layer of CNNs for speech synthesis encodes linguistically relevant information. We propose two techniques for exploration of the fully connected layer. In Experiment 1, we use weight matrices as inputs into convolutional layers. In Experiment 2, we manipulate the FC layer to explore how symbolic-like representations are encoded in CNNs. We leverage the fact that the FC layer outputs a feature map and that variable-specific weight matrices are temporally structured to (1) demonstrate how the distribution of learned weights varies between latent variables in systematic ways and (2) demonstrate how manipulating the FC layer while holding constant subsequent model parameters affects the output. We ultimately present an FC manipulation that can output a single segment. Using this technique, we show that lexically specific latent codes in generative CNNs (ciwGAN) have shared lexically invariant sublexical representations in the FC-layer weights, showing that ciwGAN encodes lexical information in a linguistically principled manner.
☆ ESURF: Simple and Effective EDU Segmentation
Segmenting text into Elemental Discourse Units (EDUs) is a fundamental task in discourse parsing. We present a new simple method for identifying EDU boundaries, and hence segmenting them, based on lexical and character n-gram features, using random forest classification. We show that the method, despite its simplicity, outperforms other methods both for segmentation and within a state of the art discourse parser. This indicates the importance of such features for identifying basic discourse elements, pointing towards potentially more training-efficient methods for discourse analysis.
♻ ☆ Quilt-1M: One Million Image-Text Pairs for Histopathology
Recent accelerations in multi-modal applications have been made possible with the plethora of image and text data available online. However, the scarcity of analogous data in the medical field, specifically in histopathology, has slowed comparable progress. To enable similar representation learning for histopathology, we turn to YouTube, an untapped resource of videos, offering $1,087$ hours of valuable educational histopathology videos from expert clinicians. From YouTube, we curate QUILT: a large-scale vision-language dataset consisting of $802, 144$ image and text pairs. QUILT was automatically curated using a mixture of models, including large language models, handcrafted algorithms, human knowledge databases, and automatic speech recognition. In comparison, the most comprehensive datasets curated for histopathology amass only around $200$K samples. We combine QUILT with datasets from other sources, including Twitter, research papers, and the internet in general, to create an even larger dataset: QUILT-1M, with $1$M paired image-text samples, marking it as the largest vision-language histopathology dataset to date. We demonstrate the value of QUILT-1M by fine-tuning a pre-trained CLIP model. Our model outperforms state-of-the-art models on both zero-shot and linear probing tasks for classifying new histopathology images across $13$ diverse patch-level datasets of $8$ different sub-pathologies and cross-modal retrieval tasks.
♻ ☆ Controlling Equational Reasoning in Large Language Models with Prompt Interventions AAAI 2025
This paper investigates how hallucination rates in Large Language Models (LLMs) may be controlled via a symbolic data generation framework, exploring a fundamental relationship between the rate of certain mathematical errors and types of input intervention. Specifically, we systematically generate data for a derivation generation task using a symbolic engine, applying targeted interventions to prompts to perturb features of mathematical derivations such as the surface forms of symbols, equational tree structures, and mathematical context. We then evaluate the effect of prompt interventions across a range of LLMs including fine-tuned T5 models, GPT, and LLaMa-based models. Our experiments suggest that T5-Large can outperform the few-shot performance of GPT-4 on various evaluation sets generated via the framework. However, an extensive evaluation based on human analysis, template-based error detection, and text generation metrics reveals model weaknesses beyond what the reference-based metrics singularly describe. We use these results to tie characteristic distributional footprints of interventions to the human evaluation of LLM derivation quality, potentially leading to significant control over fine-grained mathematical capabilities of language models with respect to specific types of errors.
comment: AAAI 2025 (7 pages)
♻ ☆ Small Language Models can Outperform Humans in Short Creative Writing: A Study Comparing SLMs with Humans and LLMs COLING 2025
In this paper, we evaluate the creative fiction writing abilities of a fine-tuned small language model (SLM), BART-large, and compare its performance to human writers and two large language models (LLMs): GPT-3.5 and GPT-4o. Our evaluation consists of two experiments: (i) a human study in which 68 participants rated short stories from humans and the SLM on grammaticality, relevance, creativity, and attractiveness, and (ii) a qualitative linguistic analysis examining the textual characteristics of stories produced by each model. In the first experiment, BART-large outscored average human writers overall (2.11 vs. 1.85), a 14% relative improvement, though the slight human advantage in creativity was not statistically significant. In the second experiment, qualitative analysis showed that while GPT-4o demonstrated near-perfect coherence and used less cliche phrases, it tended to produce more predictable language, with only 3% of its synopses featuring surprising associations (compared to 15% for BART). These findings highlight how model size and fine-tuning influence the balance between creativity, fluency, and coherence in creative writing tasks, and demonstrate that smaller models can, in certain contexts, rival both humans and larger models.
comment: Accepted as Main Conference Paper at COLING 2025
♻ ☆ Are queries and keys always relevant? A case study on Transformer wave functions
The dot product attention mechanism, originally designed for natural language processing tasks, is a cornerstone of modern Transformers. It adeptly captures semantic relationships between word pairs in sentences by computing a similarity overlap between queries and keys. In this work, we explore the suitability of Transformers, focusing on their attention mechanisms, in the specific domain of the parametrization of variational wave functions to approximate ground states of quantum many-body spin Hamiltonians. Specifically, we perform numerical simulations on the two-dimensional $J_1$-$J_2$ Heisenberg model, a common benchmark in the field of quantum many-body systems on lattice. By comparing the performance of standard attention mechanisms with a simplified version that excludes queries and keys, relying solely on positions, we achieve competitive results while reducing computational cost and parameter usage. Furthermore, through the analysis of the attention maps generated by standard attention mechanisms, we show that the attention weights become effectively input-independent at the end of the optimization. We support the numerical results with analytical calculations, providing physical insights of why queries and keys should be, in principle, omitted from the attention mechanism when studying large systems.
comment: 10 pages, 5 figures, 1 table
♻ ☆ Eradicating Social Biases in Sentiment Analysis using Semantic Blinding and Semantic Propagation Graph Neural Networks
This paper introduces the Semantic Propagation Graph Neural Network (SProp GNN), a machine learning sentiment analysis (SA) architecture that relies exclusively on syntactic structures and word-level emotional cues to predict emotions in text. By semantically blinding the model to information about specific words, it is robust to social biases such as political or gender bias that have been plaguing previous machine learning-based SA systems. The SProp GNN shows performance superior to lexicon-based alternatives such as VADER (Valence Aware Dictionary and Sentiment Reasoner) and EmoAtlas on two different prediction tasks, and across two languages. Additionally, it approaches the accuracy of transformer-based models while significantly reducing bias in emotion prediction tasks. By offering improved explainability and reducing bias, the SProp GNN bridges the methodological gap between interpretable lexicon approaches and powerful, yet often opaque, deep learning models, offering a robust tool for fair and effective emotion analysis in understanding human behavior through text.
♻ ☆ Mitigating Out-of-Entity Errors in Named Entity Recognition: A Sentence-Level Strategy COLING 2025
Many previous models of named entity recognition (NER) suffer from the problem of Out-of-Entity (OOE), i.e., the tokens in the entity mentions of the test samples have not appeared in the training samples, which hinders the achievement of satisfactory performance. To improve OOE-NER performance, in this paper, we propose a new framework, namely S+NER, which fully leverages sentence-level information. Our S+NER achieves better OOE-NER performance mainly due to the following two particular designs. 1) It first exploits the pre-trained language model's capability of understanding the target entity's sentence-level context with a template set. 2) Then, it refines the sentence-level representation based on the positive and negative templates, through a contrastive learning strategy and template pooling method, to obtain better NER results. Our extensive experiments on five benchmark datasets have demonstrated that, our S+NER outperforms some state-of-the-art OOE-NER models.
comment: Accepted by COLING 2025
♻ ☆ Don't Command, Cultivate: An Exploratory Study of System-2 Alignment
The o1 system card identifies the o1 models as the most robust within OpenAI, with their defining characteristic being the progression from rapid, intuitive thinking to slower, more deliberate reasoning. This observation motivated us to investigate the influence of System-2 thinking patterns on model safety. In our preliminary research, we conducted safety evaluations of the o1 model, including complex jailbreak attack scenarios using adversarial natural language prompts and mathematical encoding prompts. Our findings indicate that the o1 model demonstrates relatively improved safety performance; however, it still exhibits vulnerabilities, particularly against jailbreak attacks employing mathematical encoding. Through detailed case analysis, we identified specific patterns in the o1 model's responses. We also explored the alignment of System-2 safety in open-source models using prompt engineering and supervised fine-tuning techniques. Experimental results show that some simple methods to encourage the model to carefully scrutinize user requests are beneficial for model safety. Additionally, we proposed a implementation plan for process supervision to enhance safety alignment. The implementation details and experimental results will be provided in future versions.
comment: In this version, the DPO and reinforcement learning methods have been added
♻ ☆ Are LLMs Good Cryptic Crossword Solvers?
Cryptic crosswords are puzzles that rely not only on general knowledge but also on the solver's ability to manipulate language on different levels and deal with various types of wordplay. Previous research suggests that solving such puzzles is a challenge even for modern NLP models. However, the abilities of large language models (LLMs) have not yet been tested on this task. In this paper, we establish the benchmark results for three popular LLMs -- LLaMA2, Mistral, and ChatGPT -- showing that their performance on this task is still far from that of humans.
♻ ☆ VaeDiff-DocRE: End-to-end Data Augmentation Framework for Document-level Relation Extraction COLING 2025
Document-level Relation Extraction (DocRE) aims to identify relationships between entity pairs within a document. However, most existing methods assume a uniform label distribution, resulting in suboptimal performance on real-world, imbalanced datasets. To tackle this challenge, we propose a novel data augmentation approach using generative models to enhance data from the embedding space. Our method leverages the Variational Autoencoder (VAE) architecture to capture all relation-wise distributions formed by entity pair representations and augment data for underrepresented relations. To better capture the multi-label nature of DocRE, we parameterize the VAE's latent space with a Diffusion Model. Additionally, we introduce a hierarchical training framework to integrate the proposed VAE-based augmentation module into DocRE systems. Experiments on two benchmark datasets demonstrate that our method outperforms state-of-the-art models, effectively addressing the long-tail distribution problem in DocRE.
comment: COLING 2025
♻ ☆ ConSim: Measuring Concept-Based Explanations' Effectiveness with Automated Simulatability
Concept-based explanations work by mapping complex model computations to human-understandable concepts. Evaluating such explanations is very difficult, as it includes not only the quality of the induced space of possible concepts but also how effectively the chosen concepts are communicated to users. Existing evaluation metrics often focus solely on the former, neglecting the latter. We introduce an evaluation framework for measuring concept explanations via automated simulatability: a simulator's ability to predict the explained model's outputs based on the provided explanations. This approach accounts for both the concept space and its interpretation in an end-to-end evaluation. Human studies for simulatability are notoriously difficult to enact, particularly at the scale of a wide, comprehensive empirical evaluation (which is the subject of this work). We propose using large language models (LLMs) as simulators to approximate the evaluation and report various analyses to make such approximations reliable. Our method allows for scalable and consistent evaluation across various models and datasets. We report a comprehensive empirical evaluation using this framework and show that LLMs provide consistent rankings of explanation methods. Code available at https://github.com/AnonymousConSim/ConSim.
♻ ☆ Migician: Revealing the Magic of Free-Form Multi-Image Grounding in Multimodal Large Language Models
The recent advancement of Multimodal Large Language Models (MLLMs) has significantly improved their fine-grained perception of single images and general comprehension across multiple images. However, existing MLLMs still face challenges in achieving precise grounding in complex multi-image scenarios. To address this, we first explore a Chain-of-Thought (CoT) framework that integrates single-image grounding with multi-image comprehension. While partially effective, it remains unstable and struggles to capture abstract visual information due to its non-end-to-end nature. Therefore, we introduce Migician, the first multi-image grounding model capable of performing free-form and accurate grounding across multiple images. To support this, we present the MGrounding-630k dataset, which comprises data for several multi-image grounding tasks derived from existing datasets, along with newly generated free-form grounding instruction-following data. Furthermore, we propose MIG-Bench, a comprehensive benchmark specifically designed for evaluating multi-image grounding capabilities. Experimental results demonstrate that our model achieves significantly superior multi-image grounding capabilities, outperforming the best existing MLLMs by 21.61% and even surpassing much larger 70B models. Our code, model, dataset, and benchmark are fully open-sourced at https://migician-vg.github.io/.
comment: 20 pages, 8 figures
♻ ☆ DefVerify: Do Hate Speech Models Reflect Their Dataset's Definition? COLING 2025
When building a predictive model, it is often difficult to ensure that application-specific requirements are encoded by the model that will eventually be deployed. Consider researchers working on hate speech detection. They will have an idea of what is considered hate speech, but building a model that reflects their view accurately requires preserving those ideals throughout the workflow of data set construction and model training. Complications such as sampling bias, annotation bias, and model misspecification almost always arise, possibly resulting in a gap between the application specification and the model's actual behavior upon deployment. To address this issue for hate speech detection, we propose DefVerify: a 3-step procedure that (i) encodes a user-specified definition of hate speech, (ii) quantifies to what extent the model reflects the intended definition, and (iii) tries to identify the point of failure in the workflow. We use DefVerify to find gaps between definition and model behavior when applied to six popular hate speech benchmark datasets.
comment: Camera-ready COLING 2025
♻ ☆ Assessment and manipulation of latent constructs in pre-trained language models using psychometric scales
Human-like personality traits have recently been discovered in large language models, raising the hypothesis that their (known and as yet undiscovered) biases conform with human latent psychological constructs. While large conversational models may be tricked into answering psychometric questionnaires, the latent psychological constructs of thousands of simpler transformers, trained for other tasks, cannot be assessed because appropriate psychometric methods are currently lacking. Here, we show how standard psychological questionnaires can be reformulated into natural language inference prompts, and we provide a code library to support the psychometric assessment of arbitrary models. We demonstrate, using a sample of 88 publicly available models, the existence of human-like mental health-related constructs (including anxiety, depression, and Sense of Coherence) which conform with standard theories in human psychology and show similar correlations and mitigation strategies. The ability to interpret and rectify the performance of language models by using psychological tools can boost the development of more explainable, controllable, and trustworthy models.
♻ ☆ II-Bench: An Image Implication Understanding Benchmark for Multimodal Large Language Models
The rapid advancements in the development of multimodal large language models (MLLMs) have consistently led to new breakthroughs on various benchmarks. In response, numerous challenging and comprehensive benchmarks have been proposed to more accurately assess the capabilities of MLLMs. However, there is a dearth of exploration of the higher-order perceptual capabilities of MLLMs. To fill this gap, we propose the Image Implication understanding Benchmark, II-Bench, which aims to evaluate the model's higher-order perception of images. Through extensive experiments on II-Bench across multiple MLLMs, we have made significant findings. Initially, a substantial gap is observed between the performance of MLLMs and humans on II-Bench. The pinnacle accuracy of MLLMs attains 74.8%, whereas human accuracy averages 90%, peaking at an impressive 98%. Subsequently, MLLMs perform worse on abstract and complex images, suggesting limitations in their ability to understand high-level semantics and capture image details. Finally, it is observed that most models exhibit enhanced accuracy when image sentiment polarity hints are incorporated into the prompts. This observation underscores a notable deficiency in their inherent understanding of image sentiment. We believe that II-Bench will inspire the community to develop the next generation of MLLMs, advancing the journey towards expert artificial general intelligence (AGI). II-Bench is publicly available at https://huggingface.co/datasets/m-a-p/II-Bench.
comment: 100 pages, 82 figures, add citations
♻ ☆ Quilt-LLaVA: Visual Instruction Tuning by Extracting Localized Narratives from Open-Source Histopathology Videos
Diagnosis in histopathology requires a global whole slide images (WSIs) analysis, requiring pathologists to compound evidence from different WSI patches. The gigapixel scale of WSIs poses a challenge for histopathology multi-modal models. Training multi-model models for histopathology requires instruction tuning datasets, which currently contain information for individual image patches, without a spatial grounding of the concepts within each patch and without a wider view of the WSI. Therefore, they lack sufficient diagnostic capacity for histopathology. To bridge this gap, we introduce Quilt-Instruct, a large-scale dataset of 107,131 histopathology-specific instruction question/answer pairs, grounded within diagnostically relevant image patches that make up the WSI. Our dataset is collected by leveraging educational histopathology videos from YouTube, which provides spatial localization of narrations by automatically extracting the narrators' cursor positions. Quilt-Instruct supports contextual reasoning by extracting diagnosis and supporting facts from the entire WSI. Using Quilt-Instruct, we train Quilt-LLaVA, which can reason beyond the given single image patch, enabling diagnostic reasoning across patches. To evaluate Quilt-LLaVA, we propose a comprehensive evaluation dataset created from 985 images and 1283 human-generated question-answers. We also thoroughly evaluate Quilt-LLaVA using public histopathology datasets, where Quilt-LLaVA significantly outperforms SOTA by over 10% on relative GPT-4 score and 4% and 9% on open and closed set VQA. Our code, data, and model are publicly accessible at quilt-llava.github.io.
♻ ☆ MIO: A Foundation Model on Multimodal Tokens
In this paper, we introduce MIO, a novel foundation model built on multimodal tokens, capable of understanding and generating speech, text, images, and videos in an end-to-end, autoregressive manner. While the emergence of large language models (LLMs) and multimodal large language models (MM-LLMs) propels advancements in artificial general intelligence through their versatile capabilities, they still lack true any-to-any understanding and generation. Recently, the release of GPT-4o has showcased the remarkable potential of any-to-any LLMs for complex real-world tasks, enabling omnidirectional input and output across images, speech, and text. However, it is closed-source and does not support the generation of multimodal interleaved sequences. To address this gap, we present MIO, which is trained on a mixture of discrete tokens across four modalities using causal multimodal modeling. MIO undergoes a four-stage training process: (1) alignment pre-training, (2) interleaved pre-training, (3) speech-enhanced pre-training, and (4) comprehensive supervised fine-tuning on diverse textual, visual, and speech tasks. Our experimental results indicate that MIO exhibits competitive, and in some cases superior, performance compared to previous dual-modal baselines, any-to-any model baselines, and even modality-specific baselines. Moreover, MIO demonstrates advanced capabilities inherent to its any-to-any feature, such as interleaved video-text generation, chain-of-visual-thought reasoning, visual guideline generation, instructional image editing, etc.
comment: Technical Report. Codes and models are available in https://github.com/MIO-Team/MIO
♻ ☆ DiReCT: Diagnostic Reasoning for Clinical Notes via Large Language Models
Large language models (LLMs) have recently showcased remarkable capabilities, spanning a wide range of tasks and applications, including those in the medical domain. Models like GPT-4 excel in medical question answering but may face challenges in the lack of interpretability when handling complex tasks in real clinical settings. We thus introduce the diagnostic reasoning dataset for clinical notes (DiReCT), aiming at evaluating the reasoning ability and interpretability of LLMs compared to human doctors. It contains 511 clinical notes, each meticulously annotated by physicians, detailing the diagnostic reasoning process from observations in a clinical note to the final diagnosis. Additionally, a diagnostic knowledge graph is provided to offer essential knowledge for reasoning, which may not be covered in the training data of existing LLMs. Evaluations of leading LLMs on DiReCT bring out a significant gap between their reasoning ability and that of human doctors, highlighting the critical need for models that can reason effectively in real-world clinical scenarios.
comment: 9 pages,6 figures
♻ ☆ Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability
Mathematical reasoning tasks pose significant challenges for large language models (LLMs) because they require precise logical deduction and sequence analysis. In this work, we introduce the concept of critical tokens -- elements within reasoning trajectories that significantly influence incorrect outcomes. We present a novel framework for identifying these tokens through rollout sampling and demonstrate their substantial divergence from traditional error tokens. Through extensive experiments on datasets such as GSM8K and MATH500, we show that identifying and replacing critical tokens significantly improves model accuracy. We propose an efficient methodology for pinpointing these tokens in large-scale datasets using contrastive estimation and extend this framework to enhance model training processes with direct preference optimization (DPO). Experimental results on GSM8K and MATH500 benchmarks with the widely used models Llama-3 (8B and 70B) and Deepseek-math (7B) demonstrate the effectiveness of the proposed approach, cDPO. Our results underscore the potential of leveraging critical tokens to reduce errors in reasoning tasks, advancing the development of AI systems capable of robust logical deduction. Our code, annotated datasets, and trained models are available at https://github.com/chenzhiling9954/Critical-Tokens-Matter to support and encourage future research in this promising field.
comment: Work in progress
♻ ☆ Step-by-Step Mastery: Enhancing Soft Constraint Following Ability of Large Language Models
It is crucial for large language models (LLMs) to follow instructions that involve multiple constraints. However, soft constraints are semantically related and difficult to verify through automated methods. These constraints remain a significant challenge for LLMs. To enhance the ability of LLMs to follow soft constraints, we initially design a pipeline to obtain high-quality outputs automatically. Additionally, to fully utilize the acquired data, we introduce a training paradigm based on curriculum learning. We experimentally evaluate the effectiveness of our methods in improving LLMs' soft constraint following ability and analyze the factors driving the improvements. The datasets and code are publicly available at https://github.com/Rainier-rq/FollowSoftConstraints.
♻ ☆ MLLM-CompBench: A Comparative Reasoning Benchmark for Multimodal LLMs NeurIPS 2024
The ability to compare objects, scenes, or situations is crucial for effective decision-making and problem-solving in everyday life. For instance, comparing the freshness of apples enables better choices during grocery shopping while comparing sofa designs helps optimize the aesthetics of our living space. Despite its significance, the comparative capability is largely unexplored in artificial general intelligence (AGI). In this paper, we introduce MLLM-CompBench, a benchmark designed to evaluate the comparative reasoning capability of multimodal large language models (MLLMs). MLLM-CompBench mines and pairs images through visually oriented questions covering eight dimensions of relative comparison: visual attribute, existence, state, emotion, temporality, spatiality, quantity, and quality. We curate a collection of around 40K image pairs using metadata from diverse vision datasets and CLIP similarity scores. These image pairs span a broad array of visual domains, including animals, fashion, sports, and both outdoor and indoor scenes. The questions are carefully crafted to discern relative characteristics between two images and are labeled by human annotators for accuracy and relevance. We use MLLM-CompBench to evaluate recent MLLMs, including GPT-4V(ision), Gemini-Pro, and LLaVA-1.6. Our results reveal notable shortcomings in their comparative abilities. We believe MLLM-COMPBENCH not only sheds light on these limitations but also establishes a solid foundation for future enhancements in the comparative capability of MLLMs.
comment: This paper has been accepted to NeurIPS 2024. The first two authors contributed equally to this work
♻ ☆ HADES: Hardware Accelerated Decoding for Efficient Speculation in Large Language Models
Large Language Models (LLMs) have revolutionized natural language processing by understanding and generating human-like text. However, the increasing demand for more sophisticated LLMs presents significant computational challenges due to their scale and complexity. This paper introduces Hardware Accelerated Decoding (HADES), a novel approach to enhance the performance and energy efficiency of LLMs. We address the design of an LLM accelerator with hardware-level speculative decoding support, a concept not previously explored in existing literature. Our work demonstrates how speculative decoding can significantly improve the efficiency of LLM operations, paving the way for more advanced and practical applications of these models.
comment: Accepted to ICCEA 2025
♻ ☆ Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking Multimodal Large Language Models ECCV 2024
In this paper, we study the harmlessness alignment problem of multimodal large language models (MLLMs). We conduct a systematic empirical analysis of the harmlessness performance of representative MLLMs and reveal that the image input poses the alignment vulnerability of MLLMs. Inspired by this, we propose a novel jailbreak method named HADES, which hides and amplifies the harmfulness of the malicious intent within the text input, using meticulously crafted images. Experimental results show that HADES can effectively jailbreak existing MLLMs, which achieves an average Attack Success Rate (ASR) of 90.26% for LLaVA-1.5 and 71.60% for Gemini Pro Vision. Our code and data are available at https://github.com/RUCAIBox/HADES.
comment: ECCV 2024 Oral
♻ ☆ Explainable Artificial Intelligence: A Survey of Needs, Techniques, Applications, and Future Direction
Artificial intelligence models encounter significant challenges due to their black-box nature, particularly in safety-critical domains such as healthcare, finance, and autonomous vehicles. Explainable Artificial Intelligence (XAI) addresses these challenges by providing explanations for how these models make decisions and predictions, ensuring transparency, accountability, and fairness. Existing studies have examined the fundamental concepts of XAI, its general principles, and the scope of XAI techniques. However, there remains a gap in the literature as there are no comprehensive reviews that delve into the detailed mathematical representations, design methodologies of XAI models, and other associated aspects. This paper provides a comprehensive literature review encompassing common terminologies and definitions, the need for XAI, beneficiaries of XAI, a taxonomy of XAI methods, and the application of XAI methods in different application areas. The survey is aimed at XAI researchers, XAI practitioners, AI model developers, and XAI beneficiaries who are interested in enhancing the trustworthiness, transparency, accountability, and fairness of their AI models.
♻ ☆ Sparse Attention Vectors: Generative Multimodal Model Features Are Discriminative Vision-Language Classifiers
Generative Large Multimodal Models (LMMs) like LLaVA and Qwen-VL excel at a wide variety of vision-language (VL) tasks such as image captioning or visual question answering. Despite strong performance, LMMs are not directly suited for foundational discriminative vision-language tasks (i.e., tasks requiring discrete label predictions) such as image classification and multiple-choice VQA. One key challenge in utilizing LMMs for discriminative tasks is the extraction of useful features from generative models. To overcome this issue, we propose an approach for finding features in the model's latent space to more effectively leverage LMMs for discriminative tasks. Toward this end, we present Sparse Attention Vectors (SAVs) -- a finetuning-free method that leverages sparse attention head activations (fewer than 1\% of the heads) in LMMs as strong features for VL tasks. With only few-shot examples, SAVs demonstrate state-of-the-art performance compared to a variety of few-shot and finetuned baselines on a collection of discriminative tasks. Our experiments also imply that SAVs can scale in performance with additional examples and generalize to similar tasks, establishing SAVs as both effective and robust multimodal feature representations.
♻ ☆ Divergences between Language Models and Human Brains
Do machines and humans process language in similar ways? Recent research has hinted at the affirmative, showing that human neural activity can be effectively predicted using the internal representations of language models (LMs). Although such results are thought to reflect shared computational principles between LMs and human brains, there are also clear differences in how LMs and humans represent and use language. In this work, we systematically explore the divergences between human and machine language processing by examining the differences between LM representations and human brain responses to language as measured by Magnetoencephalography (MEG) across two datasets in which subjects read and listened to narrative stories. Using an LLM-based data-driven approach, we identify two domains that LMs do not capture well: social/emotional intelligence and physical commonsense. We validate these findings with human behavioral experiments and hypothesize that the gap is due to insufficient representations of social/emotional and physical knowledge in LMs. Our results show that fine-tuning LMs on these domains can improve their alignment with human brain responses.
Machine Learning 133
☆ E2ESlack: An End-to-End Graph-Based Framework for Pre-Routing Slack Prediction
Pre-routing slack prediction remains a critical area of research in Electronic Design Automation (EDA). Despite numerous machine learning-based approaches targeting this task, there is still a lack of a truly end-to-end framework that engineers can use to obtain TNS/WNS metrics from raw circuit data at the placement stage. Existing works have demonstrated effectiveness in Arrival Time (AT) prediction but lack a mechanism for Required Arrival Time (RAT) prediction, which is essential for slack prediction and obtaining TNS/WNS metrics. In this work, we propose E2ESlack, an end-to-end graph-based framework for pre-routing slack prediction. The framework includes a TimingParser that supports DEF, SDF and LIB files for feature extraction and graph construction, an arrival time prediction model and a fast RAT estimation module. To the best of our knowledge, this is the first work capable of predicting path-level slacks at the pre-routing stage. We perform extensive experiments and demonstrate that our proposed RAT estimation method outperforms the SOTA ML-based prediction method and also pre-routing STA tool. Additionally, the proposed E2ESlack framework achieves TNS/WNS values comparable to post-routing STA results while saving up to 23x runtime.
☆ Dynamic Prototype Rehearsal for Continual Learning in ECG Arrhythmia Detection ICASSP 2025
Continual Learning (CL) methods aim to learn from a sequence of tasks while avoiding the challenge of forgetting previous knowledge. We present DREAM-CL, a novel CL method for ECG arrhythmia detection that introduces dynamic prototype rehearsal memory. DREAM-CL selects representative prototypes by clustering data based on learning behavior during each training session. Within each cluster, we apply a smooth sorting operation that ranks samples by training difficulty, compressing extreme values and removing outliers. The more challenging samples are then chosen as prototypes for the rehearsal memory, ensuring effective knowledge retention across sessions. We evaluate our method on time-incremental, class-incremental, and lead-incremental scenarios using two widely used ECG arrhythmia datasets, Chapman and PTB-XL. The results demonstrate that DREAM-CL outperforms the state-of-the-art in CL for ECG arrhythmia detection. Detailed ablation and sensitivity studies are performed to validate the different design choices of our method.
comment: Accepted to 2025 International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025)
☆ Imagine while Reasoning in Space: Multimodal Visualization-of-Thought
Chain-of-Thought (CoT) prompting has proven highly effective for enhancing complex reasoning in Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs). Yet, it struggles in complex spatial reasoning tasks. Nonetheless, human cognition extends beyond language alone, enabling the remarkable capability to think in both words and images. Inspired by this mechanism, we propose a new reasoning paradigm, Multimodal Visualization-of-Thought (MVoT). It enables visual thinking in MLLMs by generating image visualizations of their reasoning traces. To ensure high-quality visualization, we introduce token discrepancy loss into autoregressive MLLMs. This innovation significantly improves both visual coherence and fidelity. We validate this approach through several dynamic spatial reasoning tasks. Experimental results reveal that MVoT demonstrates competitive performance across tasks. Moreover, it exhibits robust and reliable improvements in the most challenging scenarios where CoT fails. Ultimately, MVoT establishes new possibilities for complex reasoning tasks where visual thinking can effectively complement verbal reasoning.
comment: 11 pages, 6 figures, 4 tables (27 pages, 10 figures, 16 tables including references and appendices)
☆ ML Mule: Mobile-Driven Context-Aware Collaborative Learning
Artificial intelligence has been integrated into nearly every aspect of daily life, powering applications from object detection with computer vision to large language models for writing emails and compact models in smart homes. These machine learning models cater to individual users but are often detached from them, as they are typically stored and processed in centralized data centers. This centralized approach raises privacy concerns, incurs high infrastructure costs, and struggles with personalization. Federated and fully decentralized learning methods have been proposed to address these issues, but they still depend on centralized servers or face slow convergence due to communication constraints. To overcome these challenges, we propose ML Mule, a approach that utilizes individual mobile devices as 'Mules' to train and transport model snapshots as they move through physical spaces, sharing these models with the physical 'Spaces' they inhabit. This method implicitly forms affinity groups among devices associated with users who share particular spaces, enabling collaborative model evolution, and protecting users' privacy. Our approach addresses several major shortcomings of traditional, federated, and fully decentralized learning systems. The proposed framework represents a new class of machine learning methods that are more robust, distributed, and personalized, bringing the field closer to realizing the original vision of intelligent, adaptive, and genuinely context-aware smart environments. The results show that ML Mule converges faster and achieves higher model accuracy compared to other existing methods.
☆ Investigating Map-Based Path Loss Models: A Study of Feature Representations in Convolutional Neural Networks
Path loss prediction is a beneficial tool for efficient use of the radio frequency spectrum. Building on prior research on high-resolution map-based path loss models, this paper studies convolutional neural network input representations in more detail. We investigate different methods of representing scalar features in convolutional neural networks. Specifically, we compare using frequency and distance as input channels to convolutional layers or as scalar inputs to regression layers. We assess model performance using three different feature configurations and find that representing scalar features as image channels results in the strongest generalization.
comment: 4 pages, 2 figures, 4 tables
☆ RadAlign: Advancing Radiology Report Generation with Vision-Language Concept Alignment
Automated chest radiographs interpretation requires both accurate disease classification and detailed radiology report generation, presenting a significant challenge in the clinical workflow. Current approaches either focus on classification accuracy at the expense of interpretability or generate detailed but potentially unreliable reports through image captioning techniques. In this study, we present RadAlign, a novel framework that combines the predictive accuracy of vision-language models (VLMs) with the reasoning capabilities of large language models (LLMs). Inspired by the radiologist's workflow, RadAlign first employs a specialized VLM to align visual features with key medical concepts, achieving superior disease classification with an average AUC of 0.885 across multiple diseases. These recognized medical conditions, represented as text-based concepts in the aligned visual-language space, are then used to prompt LLM-based report generation. Enhanced by a retrieval-augmented generation mechanism that grounds outputs in similar historical cases, RadAlign delivers superior report quality with a GREEN score of 0.678, outperforming state-of-the-art methods' 0.634. Our framework maintains strong clinical interpretability while reducing hallucinations, advancing automated medical imaging and report analysis through integrated predictive and generative AI. Code is available at https://github.com/difeigu/RadAlign.
☆ Improving DeFi Accessibility through Efficient Liquidity Provisioning with Deep Reinforcement Learning AAAI 2025
This paper applies deep reinforcement learning (DRL) to optimize liquidity provisioning in Uniswap v3, a decentralized finance (DeFi) protocol implementing an automated market maker (AMM) model with concentrated liquidity. We model the liquidity provision task as a Markov Decision Process (MDP) and train an active liquidity provider (LP) agent using the Proximal Policy Optimization (PPO) algorithm. The agent dynamically adjusts liquidity positions by using information about price dynamics to balance fee maximization and impermanent loss mitigation. We use a rolling window approach for training and testing, reflecting realistic market conditions and regime shifts. This study compares the data-driven performance of the DRL-based strategy against common heuristics adopted by small retail LP actors that do not systematically modify their liquidity positions. By promoting more efficient liquidity management, this work aims to make DeFi markets more accessible and inclusive for a broader range of participants. Through a data-driven approach to liquidity management, this work seeks to contribute to the ongoing development of more efficient and user-friendly DeFi markets.
comment: 9 pages, 5 figures. Accepted at AI for Social Impact: Bridging Innovations in Finance, Social Media, and Crime Prevention Workshop at AAAI 2025
☆ RbRL2.0: Integrated Reward and Policy Learning for Rating-based Reinforcement Learning AAAI 2025
Reinforcement learning (RL), a common tool in decision making, learns policies from various experiences based on the associated cumulative return/rewards without treating them differently. On the contrary, humans often learn to distinguish from different levels of performance and extract the underlying trends towards improving their decision making for best performance. Motivated by this, this paper proposes a novel RL method that mimics humans' decision making process by differentiating among collected experiences for effective policy learning. The main idea is to extract important directional information from experiences with different performance levels, named ratings, so that policies can be updated towards desired deviation from these experiences with different ratings. Specifically, we propose a new policy loss function that penalizes distribution similarities between the current policy and failed experiences with different ratings, and assign different weights to the penalty terms based on the rating classes. Meanwhile, reward learning from these rated samples can be integrated with the new policy loss towards an integrated reward and policy learning from rated samples. Optimizing the integrated reward and policy loss function will lead to the discovery of directions for policy improvement towards maximizing cumulative rewards and penalizing most from the lowest performance level while least from the highest performance level. To evaluate the effectiveness of the proposed method, we present results for experiments on a few typical environments that show improved convergence and overall performance over the existing rating-based reinforcement learning method with only reward learning.
comment: Accepted to the Collaborative AI and Modeling of Humans Bridge Program at AAAI 2025
☆ Exploring and Mitigating Adversarial Manipulation of Voting-Based Leaderboards
It is now common to evaluate Large Language Models (LLMs) by having humans manually vote to evaluate model outputs, in contrast to typical benchmarks that evaluate knowledge or skill at some particular task. Chatbot Arena, the most popular benchmark of this type, ranks models by asking users to select the better response between two randomly selected models (without revealing which model was responsible for the generations). These platforms are widely trusted as a fair and accurate measure of LLM capabilities. In this paper, we show that if bot protection and other defenses are not implemented, these voting-based benchmarks are potentially vulnerable to adversarial manipulation. Specifically, we show that an attacker can alter the leaderboard (to promote their favorite model or demote competitors) at the cost of roughly a thousand votes (verified in a simulated, offline version of Chatbot Arena). Our attack consists of two steps: first, we show how an attacker can determine which model was used to generate a given reply with more than $95\%$ accuracy; and then, the attacker can use this information to consistently vote for (or against) a target model. Working with the Chatbot Arena developers, we identify, propose, and implement mitigations to improve the robustness of Chatbot Arena against adversarial manipulation, which, based on our analysis, substantially increases the cost of such attacks. Some of these defenses were present before our collaboration, such as bot protection with Cloudflare, malicious user detection, and rate limiting. Others, including reCAPTCHA and login are being integrated to strengthen the security in Chatbot Arena.
☆ PrecipDiff: Leveraging image diffusion models to enhance satellite-based precipitation observations
A recent report from the World Meteorological Organization (WMO) highlights that water-related disasters have caused the highest human losses among natural disasters over the past 50 years, with over 91\% of deaths occurring in low-income countries. This disparity is largely due to the lack of adequate ground monitoring stations, such as weather surveillance radars (WSR), which are expensive to install. For example, while the US and Europe combined possess over 600 WSRs, Africa, despite having almost one and half times their landmass, has fewer than 40. To address this issue, satellite-based observations offer a global, near-real-time monitoring solution. However, they face several challenges like accuracy, bias, and low spatial resolution. This study leverages the power of diffusion models and residual learning to address these limitations in a unified framework. We introduce the first diffusion model for correcting the inconsistency between different precipitation products. Our method demonstrates the effectiveness in downscaling satellite precipitation estimates from 10 km to 1 km resolution. Extensive experiments conducted in the Seattle region demonstrate significant improvements in accuracy, bias reduction, and spatial detail. Importantly, our approach achieves these results using only precipitation data, showcasing the potential of a purely computer vision-based approach for enhancing satellite precipitation products and paving the way for further advancements in this domain.
☆ Synthesis and Analysis of Data as Probability Measures with Entropy-Regularized Optimal Transport
We consider synthesis and analysis of probability measures using the entropy-regularized Wasserstein-2 cost and its unbiased version, the Sinkhorn divergence. The synthesis problem consists of computing the barycenter, with respect to these costs, of $m$ reference measures given a set of coefficients belonging to the $m$-dimensional simplex. The analysis problem consists of finding the coefficients for the closest barycenter in the Wasserstein-2 distance to a given measure $\mu$. Under the weakest assumptions on the measures thus far in the literature, we compute the derivative of the entropy-regularized Wasserstein-2 cost. We leverage this to establish a characterization of regularized barycenters as solutions to a fixed-point equation for the average of the entropic maps from the barycenter to the reference measures. This characterization yields a finite-dimensional, convex, quadratic program for solving the analysis problem when $\mu$ is a barycenter. It is shown that these coordinates, as well as the value of the barycenter functional, can be estimated from samples with dimension-independent rates of convergence, a hallmark of entropy-regularized optimal transport, and we verify these rates experimentally. We also establish that barycentric coordinates are stable with respect to perturbations in the Wasserstein-2 metric, suggesting a robustness of these coefficients to corruptions. We employ the barycentric coefficients as features for classification of corrupted point cloud data, and show that compared to neural network baselines, our approach is more efficient in small training data regimes.
comment: 58 pages. Code to reproduce experiments: https://github.com/brendanmallery9/Entropic-Barycenters
☆ Pairwise Comparisons without Stochastic Transitivity: Model, Theory and Applications
Most statistical models for pairwise comparisons, including the Bradley-Terry (BT) and Thurstone models and many extensions, make a relatively strong assumption of stochastic transitivity. This assumption imposes the existence of an unobserved global ranking among all the players/teams/items and monotone constraints on the comparison probabilities implied by the global ranking. However, the stochastic transitivity assumption does not hold in many real-world scenarios of pairwise comparisons, especially games involving multiple skills or strategies. As a result, models relying on this assumption can have suboptimal predictive performance. In this paper, we propose a general family of statistical models for pairwise comparison data without a stochastic transitivity assumption, substantially extending the BT and Thurstone models. In this model, the pairwise probabilities are determined by a (approximately) low-dimensional skew-symmetric matrix. Likelihood-based estimation methods and computational algorithms are developed, which allow for sparse data with only a small proportion of observed pairs. Theoretical analysis shows that the proposed estimator achieves minimax-rate optimality, which adapts effectively to the sparsity level of the data. The spectral theory for skew-symmetric matrices plays a crucial role in the implementation and theoretical analysis. The proposed method's superiority against the BT model, along with its broad applicability across diverse scenarios, is further supported by simulations and real data analysis.
comment: 34 pages, 1 figure
☆ Distance Measure Based on an Embedding of the Manifold of K-Component Gaussian Mixture Models into the Manifold of Symmetric Positive Definite Matrices
In this paper, a distance between the Gaussian Mixture Models(GMMs) is obtained based on an embedding of the K-component Gaussian Mixture Model into the manifold of the symmetric positive definite matrices. Proof of embedding of K-component GMMs into the manifold of symmetric positive definite matrices is given and shown that it is a submanifold. Then, proved that the manifold of GMMs with the pullback of induced metric is isometric to the submanifold with the induced metric. Through this embedding we obtain a general lower bound for the Fisher-Rao metric. This lower bound is a distance measure on the manifold of GMMs and we employ it for the similarity measure of GMMs. The effectiveness of this framework is demonstrated through an experiment on standard machine learning benchmarks, achieving accuracy of 98%, 92%, and 93.33% on the UIUC, KTH-TIPS, and UMD texture recognition datasets respectively.
☆ MVICAD2: Multi-View Independent Component Analysis with Delays and Dilations
Machine learning techniques in multi-view settings face significant challenges, particularly when integrating heterogeneous data, aligning feature spaces, and managing view-specific biases. These issues are prominent in neuroscience, where data from multiple subjects exposed to the same stimuli are analyzed to uncover brain activity dynamics. In magnetoencephalography (MEG), where signals are captured at the scalp level, estimating the brain's underlying sources is crucial, especially in group studies where sources are assumed to be similar for all subjects. Common methods, such as Multi-View Independent Component Analysis (MVICA), assume identical sources across subjects, but this assumption is often too restrictive due to individual variability and age-related changes. Multi-View Independent Component Analysis with Delays (MVICAD) addresses this by allowing sources to differ up to a temporal delay. However, temporal dilation effects, particularly in auditory stimuli, are common in brain dynamics, making the estimation of time delays alone insufficient. To address this, we propose Multi-View Independent Component Analysis with Delays and Dilations (MVICAD2), which allows sources to differ across subjects in both temporal delays and dilations. We present a model with identifiable sources, derive an approximation of its likelihood in closed form, and use regularization and optimization techniques to enhance performance. Through simulations, we demonstrate that MVICAD2 outperforms existing multi-view ICA methods. We further validate its effectiveness using the Cam-CAN dataset, and showing how delays and dilations are related to aging.
comment: 19 pages, 8 figures
☆ An Investigation into Seasonal Variations in Energy Forecasting for Student Residences
This research provides an in-depth evaluation of various machine learning models for energy forecasting, focusing on the unique challenges of seasonal variations in student residential settings. The study assesses the performance of baseline models, such as LSTM and GRU, alongside state-of-the-art forecasting methods, including Autoregressive Feedforward Neural Networks, Transformers, and hybrid approaches. Special attention is given to predicting energy consumption amidst challenges like seasonal patterns, vacations, meteorological changes, and irregular human activities that cause sudden fluctuations in usage. The findings reveal that no single model consistently outperforms others across all seasons, emphasizing the need for season-specific model selection or tailored designs. Notably, the proposed Hyper Network based LSTM and MiniAutoEncXGBoost models exhibit strong adaptability to seasonal variations, effectively capturing abrupt changes in energy consumption during summer months. This study advances the energy forecasting field by emphasizing the critical role of seasonal dynamics and model-specific behavior in achieving accurate predictions.
☆ PROTECT: Protein circadian time prediction using unsupervised learning
Circadian rhythms regulate the physiology and behavior of humans and animals. Despite advancements in understanding these rhythms and predicting circadian phases at the transcriptional level, predicting circadian phases from proteomic data remains elusive. This challenge is largely due to the scarcity of time labels in proteomic datasets, which are often characterized by small sample sizes, high dimensionality, and significant noise. Furthermore, existing methods for predicting circadian phases from transcriptomic data typically rely on prior knowledge of known rhythmic genes, making them unsuitable for proteomic datasets. To address this gap, we developed a novel computational method using unsupervised deep learning techniques to predict circadian sample phases from proteomic data without requiring time labels or prior knowledge of proteins or genes. Our model involves a two-stage training process optimized for robust circadian phase prediction: an initial greedy one-layer-at-a-time pre-training which generates informative initial parameters followed by fine-tuning. During fine-tuning, a specialized loss function guides the model to align protein expression levels with circadian patterns, enabling it to accurately capture the underlying rhythmic structure within the data. We tested our method on both time-labeled and unlabeled proteomic data. For labeled data, we compared our predictions to the known time labels, achieving high accuracy, while for unlabeled human datasets, including postmortem brain regions and urine samples, we explored circadian disruptions. Notably, our analysis identified disruptions in rhythmic proteins between Alzheimer's disease and control subjects across these samples.
☆ Derivation of effective gradient flow equations and dynamical truncation of training data in Deep Learning
We derive explicit equations governing the cumulative biases and weights in Deep Learning with ReLU activation function, based on gradient descent for the Euclidean cost in the input layer, and under the assumption that the weights are, in a precise sense, adapted to the coordinate system distinguished by the activations. We show that gradient descent corresponds to a dynamical process in the input layer, whereby clusters of data are progressively reduced in complexity ("truncated") at an exponential rate that increases with the number of data points that have already been truncated. We provide a detailed discussion of several types of solutions to the gradient flow equations. A main motivation for this work is to shed light on the interpretability question in supervised learning.
comment: AMS Latex, 35 pages
☆ Information-Theoretic Dual Memory System for Continual Learning
Continuously acquiring new knowledge from a dynamic environment is a fundamental capability for animals, facilitating their survival and ability to address various challenges. This capability is referred to as continual learning, which focuses on the ability to learn a sequence of tasks without the detriment of previous knowledge. A prevalent strategy to tackle continual learning involves selecting and storing numerous essential data samples from prior tasks within a fixed-size memory buffer. However, the majority of current memory-based techniques typically utilize a single memory buffer, which poses challenges in concurrently managing newly acquired and previously learned samples. Drawing inspiration from the Complementary Learning Systems (CLS) theory, which defines rapid and gradual learning mechanisms for processing information, we propose an innovative dual memory system called the Information-Theoretic Dual Memory System (ITDMS). This system comprises a fast memory buffer designed to retain temporary and novel samples, alongside a slow memory buffer dedicated to preserving critical and informative samples. The fast memory buffer is optimized employing an efficient reservoir sampling process. Furthermore, we introduce a novel information-theoretic memory optimization strategy that selectively identifies and retains diverse and informative data samples for the slow memory buffer. Additionally, we propose a novel balanced sample selection procedure that automatically identifies and eliminates redundant memorized samples, thus freeing up memory capacity for new data acquisitions, which can deal with a growing array of tasks. Our methodology is rigorously assessed through a series of continual learning experiments, with empirical results underscoring the effectiveness of the proposed system.
comment: 35 pages, 9 figures, submitted to Knowledge-Based Systems
☆ Dynami-CAL GraphNet: A Physics-Informed Graph Neural Network Conserving Linear and Angular Momentum for Dynamical Systems
Accurate, interpretable, and real-time modeling of multi-body dynamical systems is essential for predicting behaviors and inferring physical properties in natural and engineered environments. Traditional physics-based models face scalability challenges and are computationally demanding, while data-driven approaches like Graph Neural Networks (GNNs) often lack physical consistency, interpretability, and generalization. In this paper, we propose Dynami-CAL GraphNet, a Physics-Informed Graph Neural Network that integrates the learning capabilities of GNNs with physics-based inductive biases to address these limitations. Dynami-CAL GraphNet enforces pairwise conservation of linear and angular momentum for interacting nodes using edge-local reference frames that are equivariant to rotational symmetries, invariant to translations, and equivariant to node permutations. This design ensures physically consistent predictions of node dynamics while offering interpretable, edge-wise linear and angular impulses resulting from pairwise interactions. Evaluated on a 3D granular system with inelastic collisions, Dynami-CAL GraphNet demonstrates stable error accumulation over extended rollouts, effective extrapolations to unseen configurations, and robust handling of heterogeneous interactions and external forces. Dynami-CAL GraphNet offers significant advantages in fields requiring accurate, interpretable, and real-time modeling of complex multi-body dynamical systems, such as robotics, aerospace engineering, and materials science. By providing physically consistent and scalable predictions that adhere to fundamental conservation laws, it enables the inference of forces and moments while efficiently handling heterogeneous interactions and external forces.
☆ Simulating the Hubbard Model with Equivariant Normalizing Flows
Generative models, particularly normalizing flows, have shown exceptional performance in learning probability distributions across various domains of physics, including statistical mechanics, collider physics, and lattice field theory. In the context of lattice field theory, normalizing flows have been successfully applied to accurately learn the Boltzmann distribution, enabling a range of tasks such as direct estimation of thermodynamic observables and sampling independent and identically distributed (i.i.d.) configurations. In this work, we present a proof-of-concept demonstration that normalizing flows can be used to learn the Boltzmann distribution for the Hubbard model. This model is widely employed to study the electronic structure of graphene and other carbon nanomaterials. State-of-the-art numerical simulations of the Hubbard model, such as those based on Hybrid Monte Carlo (HMC) methods, often suffer from ergodicity issues, potentially leading to biased estimates of physical observables. Our numerical experiments demonstrate that leveraging i.i.d.\ sampling from the normalizing flow effectively addresses these issues.
comment: 14 pages, 5 figures, contribution to the 41st International Symposium on Lattice Field Theory (Lattice 2024), July 28th - August 3rd, 2024, Liverpool, UK
☆ Multimodal semantic retrieval for product search
Semantic retrieval (also known as dense retrieval) based on textual data has been extensively studied for both web search and product search application fields, where the relevance of a query and a potential target document is computed by their dense vector representation comparison. Product image is crucial for e-commence search interactions and is a key factor for customers at product explorations. But its impact for semantic retrieval has not been well studied yet. In this research, we build a multimodal representation for product items in e-commerece search in contrast to pure-text representation of products, and investigate the impact of such representations. The models are developed and evaluated on e-commerce datasets. We demonstrate that a multimodal representation scheme for a product can show improvement either on purchase recall or relevance accuracy in semantic retrieval. Additionally, we provide numerical analysis for exclusive matches retrieved by a multimodal semantic retrieval model versus a text-only semantic retrieval model, to demonstrate the validation of multimodal solutions.
☆ TimberVision: A Multi-Task Dataset and Framework for Log-Component Segmentation and Tracking in Autonomous Forestry Operations WACV
Timber represents an increasingly valuable and versatile resource. However, forestry operations such as harvesting, handling and measuring logs still require substantial human labor in remote environments posing significant safety risks. Progressively automating these tasks has the potential of increasing their efficiency as well as safety, but requires an accurate detection of individual logs as well as live trees and their context. Although initial approaches have been proposed for this challenging application domain, specialized data and algorithms are still too scarce to develop robust solutions. To mitigate this gap, we introduce the TimberVision dataset, consisting of more than 2k annotated RGB images containing a total of 51k trunk components including cut and lateral surfaces, thereby surpassing any existing dataset in this domain in terms of both quantity and detail by a large margin. Based on this data, we conduct a series of ablation experiments for oriented object detection and instance segmentation and evaluate the influence of multiple scene parameters on model performance. We introduce a generic framework to fuse the components detected by our models for both tasks into unified trunk representations. Furthermore, we automatically derive geometric properties and apply multi-object tracking to further enhance robustness. Our detection and tracking approach provides highly descriptive and accurate trunk representations solely from RGB image data, even under challenging environmental conditions. Our solution is suitable for a wide range of application scenarios and can be readily combined with other sensor modalities.
comment: Accepted at Winter Conference on Applications of Computer Vision (WACV) 2025. Code and dataset available at https://github.com/timbervision/timbervision
☆ Deep Generative Clustering with VAEs and Expectation-Maximization
We propose a novel deep clustering method that integrates Variational Autoencoders (VAEs) into the Expectation-Maximization (EM) framework. Our approach models the probability distribution of each cluster with a VAE and alternates between updating model parameters by maximizing the Evidence Lower Bound (ELBO) of the log-likelihood and refining cluster assignments based on the learned distributions. This enables effective clustering and generation of new samples from each cluster. Unlike existing VAE-based methods, our approach eliminates the need for a Gaussian Mixture Model (GMM) prior or additional regularization techniques. Experiments on MNIST and FashionMNIST demonstrate superior clustering performance compared to state-of-the-art methods.
☆ Enhancing Online Reinforcement Learning with Meta-Learned Objective from Offline Data AAAI 2025
A major challenge in Reinforcement Learning (RL) is the difficulty of learning an optimal policy from sparse rewards. Prior works enhance online RL with conventional Imitation Learning (IL) via a handcrafted auxiliary objective, at the cost of restricting the RL policy to be sub-optimal when the offline data is generated by a non-expert policy. Instead, to better leverage valuable information in offline data, we develop Generalized Imitation Learning from Demonstration (GILD), which meta-learns an objective that distills knowledge from offline data and instills intrinsic motivation towards the optimal policy. Distinct from prior works that are exclusive to a specific RL algorithm, GILD is a flexible module intended for diverse vanilla off-policy RL algorithms. In addition, GILD introduces no domain-specific hyperparameter and minimal increase in computational cost. In four challenging MuJoCo tasks with sparse rewards, we show that three RL algorithms enhanced with GILD significantly outperform state-of-the-art methods.
comment: Accepted by AAAI 2025 (this version includes supplementary material)
☆ Digital Operating Mode Classification of Real-World Amateur Radio Transmissions ICASSP 2025
This study presents an ML approach for classifying digital radio operating modes evaluated on real-world transmissions. We generated 98 different parameterized radio signals from 17 digital operating modes, transmitted each of them on the 70 cm (UHF) amateur radio band, and recorded our transmissions with two different architectures of SDR receivers. Three lightweight ML models were trained exclusively on spectrograms of limited non-transmitted signals with random characters as payloads. This training involved an online data augmentation pipeline to simulate various radio channel impairments. Our best model, EfficientNetB0, achieved an accuracy of 93.80% across the 17 operating modes and 85.47% across all 98 parameterized radio signals, evaluated on our real-world transmissions with Wikipedia articles as payloads. Furthermore, we analyzed the impact of varying signal durations & the number of FFT bins on classification, assessed the effectiveness of our simulated channel impairments, and tested our models across multiple simulated SNRs.
comment: Conference IEEE ICASSP 2025
☆ TempoGPT: Enhancing Temporal Reasoning via Quantizing Embedding
Multi-modal language model has made advanced progress in vision and audio, but still faces significant challenges in dealing with complex reasoning tasks in the time series domain. The reasons are twofold. First, labels for multi-modal time series data are coarse and devoid of analysis or reasoning processes. Training with these data cannot improve the model's reasoning capabilities. Second, due to the lack of precise tokenization in processing time series, the representation patterns for temporal and textual information are inconsistent, which hampers the effectiveness of multi-modal alignment. To address these challenges, we propose a multi-modal time series data construction approach and a multi-modal time series language model (TLM), TempoGPT. Specially, we construct multi-modal data for complex reasoning tasks by analyzing the variable-system relationships within a white-box system. Additionally, proposed TempoGPT achieves consistent representation between temporal and textual information by quantizing temporal embeddings, where temporal embeddings are quantized into a series of discrete tokens using a predefined codebook; subsequently, a shared embedding layer processes both temporal and textual tokens. Extensive experiments demonstrate that TempoGPT accurately perceives temporal information, logically infers conclusions, and achieves state-of-the-art in the constructed complex time series reasoning tasks. Moreover, we quantitatively demonstrate the effectiveness of quantizing temporal embeddings in enhancing multi-modal alignment and the reasoning capabilities of TLMs. Code and data are available at https://github.com/zhanghaochuan20/TempoGPT.
☆ Foundation Models at Work: Fine-Tuning for Fairness in Algorithmic Hiring AAAI 2025
Foundation models require fine-tuning to ensure their generative outputs align with intended results for specific tasks. Automating this fine-tuning process is challenging, as it typically needs human feedback that can be expensive to acquire. We present AutoRefine, a method that leverages reinforcement learning for targeted fine-tuning, utilizing direct feedback from measurable performance improvements in specific downstream tasks. We demonstrate the method for a problem arising in algorithmic hiring platforms where linguistic biases influence a recommendation system. In this setting, a generative model seeks to rewrite given job specifications to receive more diverse candidate matches from a recommendation engine which matches jobs to candidates. Our model detects and regulates biases in job descriptions to meet diversity and fairness criteria. The experiments on a public hiring dataset and a real-world hiring platform showcase how large language models can assist in identifying and mitigation biases in the real world.
comment: Accepted to AAAI 2025, AI Governance Workshop
☆ Evaluation of Artificial Intelligence Methods for Lead Time Prediction in Non-Cycled Areas of Automotive Production
The present study examines the effectiveness of applying Artificial Intelligence methods in an automotive production environment to predict unknown lead times in a non-cycle-controlled production area. Data structures are analyzed to identify contextual features and then preprocessed using one-hot encoding. Methods selection focuses on supervised machine learning techniques. In supervised learning methods, regression and classification methods are evaluated. Continuous regression based on target size distribution is not feasible. Classification methods analysis shows that Ensemble Learning and Support Vector Machines are the most suitable. Preliminary study results indicate that gradient boosting algorithms LightGBM, XGBoost, and CatBoost yield the best results. After further testing and extensive hyperparameter optimization, the final method choice is the LightGBM algorithm. Depending on feature availability and prediction interval granularity, relative prediction accuracies of up to 90% can be achieved. Further tests highlight the importance of periodic retraining of AI models to accurately represent complex production processes using the database. The research demonstrates that AI methods can be effectively applied to highly variable production data, adding business value by providing an additional metric for various control tasks while outperforming current non AI-based systems.
comment: 7 pages, 4 figures, CLC2024 Conference
☆ Variable Bregman Majorization-Minimization Algorithm and its Application to Dirichlet Maximum Likelihood Estimation
We propose a novel Bregman descent algorithm for minimizing a convex function that is expressed as the sum of a differentiable part (defined over an open set) and a possibly nonsmooth term. The approach, referred to as the Variable Bregman Majorization-Minimization (VBMM) algorithm, extends the Bregman Proximal Gradient method by allowing the Bregman function used in the divergence to adaptively vary at each iteration, provided it satisfies a majorizing condition on the objective function. This adaptive framework enables the algorithm to approximate the objective more precisely at each iteration, thereby allowing for accelerated convergence compared to the traditional Bregman Proximal Gradient descent. We establish the convergence of the VBMM algorithm to a minimizer under mild assumptions on the family of metrics used. Furthermore, we introduce a novel application of both the Bregman Proximal Gradient method and the VBMM algorithm to the estimation of the multidimensional parameters of a Dirichlet distribution through the maximization of its log-likelihood. Numerical experiments confirm that the VBMM algorithm outperforms existing approaches in terms of convergence speed.
☆ Code and Pixels: Multi-Modal Contrastive Pre-training for Enhanced Tabular Data Analysis
Learning from tabular data is of paramount importance, as it complements the conventional analysis of image and video data by providing a rich source of structured information that is often critical for comprehensive understanding and decision-making processes. We present Multi-task Contrastive Masked Tabular Modeling (MT-CMTM), a novel method aiming to enhance tabular models by leveraging the correlation between tabular data and corresponding images. MT-CMTM employs a dual strategy combining contrastive learning with masked tabular modeling, optimizing the synergy between these data modalities. Central to our approach is a 1D Convolutional Neural Network with residual connections and an attention mechanism (1D-ResNet-CBAM), designed to efficiently process tabular data without relying on images. This enables MT-CMTM to handle purely tabular data for downstream tasks, eliminating the need for potentially costly image acquisition and processing. We evaluated MT-CMTM on the DVM car dataset, which is uniquely suited for this particular scenario, and the newly developed HIPMP dataset, which connects membrane fabrication parameters with image data. Our MT-CMTM model outperforms the proposed tabular 1D-ResNet-CBAM, which is trained from scratch, achieving a relative 1.48% improvement in relative MSE on HIPMP and a 2.38% increase in absolute accuracy on DVM. These results demonstrate MT-CMTM's robustness and its potential to advance the field of multi-modal learning.
☆ The Lessons of Developing Process Reward Models in Mathematical Reasoning
Process Reward Models (PRMs) emerge as a promising approach for process supervision in mathematical reasoning of Large Language Models (LLMs), which aim to identify and mitigate intermediate errors in the reasoning processes. However, the development of effective PRMs faces significant challenges, particularly in data annotation and evaluation methodologies. In this paper, through extensive experiments, we demonstrate that commonly used Monte Carlo (MC) estimation-based data synthesis for PRMs typically yields inferior performance and generalization compared to LLM-as-a-judge and human annotation methods. MC estimation relies on completion models to evaluate current-step correctness, leading to inaccurate step verification. Furthermore, we identify potential biases in conventional Best-of-N (BoN) evaluation strategies for PRMs: (1) The unreliable policy models generate responses with correct answers but flawed processes, leading to a misalignment between the evaluation criteria of BoN and the PRM objectives of process verification. (2) The tolerance of PRMs of such responses leads to inflated BoN scores. (3) Existing PRMs have a significant proportion of minimum scores concentrated on the final answer steps, revealing the shift from process to outcome-based assessment in BoN Optimized PRMs. To address these challenges, we develop a consensus filtering mechanism that effectively integrates MC estimation with LLM-as-a-judge and advocates a more comprehensive evaluation framework that combines response-level and step-level metrics. Based on the mechanisms, we significantly improve both model performance and data efficiency in the BoN evaluation and the step-wise error identification task. Finally, we release a new state-of-the-art PRM that outperforms existing open-source alternatives and provides practical guidelines for future research in building process supervision models.
Dataset-Agnostic Recommender Systems
[This is a position paper and does not contain any empirical or theoretical results] Recommender systems have become a cornerstone of personalized user experiences, yet their development typically involves significant manual intervention, including dataset-specific feature engineering, hyperparameter tuning, and configuration. To this end, we introduce a novel paradigm: Dataset-Agnostic Recommender Systems (DAReS) that aims to enable a single codebase to autonomously adapt to various datasets without the need for fine-tuning, for a given recommender system task. Central to this approach is the Dataset Description Language (DsDL), a structured format that provides metadata about the dataset's features and labels, and allow the system to understand dataset's characteristics, allowing it to autonomously manage processes like feature selection, missing values imputation, noise removal, and hyperparameter optimization. By reducing the need for domain-specific expertise and manual adjustments, DAReS offers a more efficient and scalable solution for building recommender systems across diverse application domains. It addresses critical challenges in the field, such as reusability, reproducibility, and accessibility for non-expert users or entry-level researchers.
☆ Estimating quantum relative entropies on quantum computers
Quantum relative entropy, a quantum generalization of the well-known Kullback-Leibler divergence, serves as a fundamental measure of the distinguishability between quantum states and plays a pivotal role in quantum information science. Despite its importance, efficiently estimating quantum relative entropy between two quantum states on quantum computers remains a significant challenge. In this work, we propose the first quantum algorithm for estimating quantum relative entropy and Petz R\'{e}nyi divergence from two unknown quantum states on quantum computers, addressing open problems highlighted in [Phys. Rev. A 109, 032431 (2024)] and [IEEE Trans. Inf. Theory 70, 5653-5680 (2024)]. This is achieved by combining quadrature approximations of relative entropies, the variational representation of quantum f-divergences, and a new technique for parameterizing Hermitian polynomial operators to estimate their traces with quantum states. Notably, the circuit size of our algorithm is at most 2n+1 with n being the number of qubits in the quantum states and it is directly applicable to distributed scenarios, where quantum states to be compared are hosted on cross-platform quantum computers. We validate our algorithm through numerical simulations, laying the groundwork for its future deployment on quantum hardware devices.
comment: 24 pages, 10 figures; comments are welcome
☆ Bridging Smart Meter Gaps: A Benchmark of Statistical, Machine Learning and Time Series Foundation Models for Data Imputation
The integrity of time series data in smart grids is often compromised by missing values due to sensor failures, transmission errors, or disruptions. Gaps in smart meter data can bias consumption analyses and hinder reliable predictions, causing technical and economic inefficiencies. As smart meter data grows in volume and complexity, conventional techniques struggle with its nonlinear and nonstationary patterns. In this context, Generative Artificial Intelligence offers promising solutions that may outperform traditional statistical methods. In this paper, we evaluate two general-purpose Large Language Models and five Time Series Foundation Models for smart meter data imputation, comparing them with conventional Machine Learning and statistical models. We introduce artificial gaps (30 minutes to one day) into an anonymized public dataset to test inference capabilities. Results show that Time Series Foundation Models, with their contextual understanding and pattern recognition, could significantly enhance imputation accuracy in certain cases. However, the trade-off between computational cost and performance gains remains a critical consideration.
☆ Generating Poisoning Attacks against Ridge Regression Models with Categorical Features
Machine Learning (ML) models have become a very powerful tool to extract information from large datasets and use it to make accurate predictions and automated decisions. However, ML models can be vulnerable to external attacks, causing them to underperform or deviate from their expected tasks. One way to attack ML models is by injecting malicious data to mislead the algorithm during the training phase, which is referred to as a poisoning attack. We can prepare for such situations by designing anticipated attacks, which are later used for creating and testing defence strategies. In this paper, we propose an algorithm to generate strong poisoning attacks for a ridge regression model containing both numerical and categorical features that explicitly models and poisons categorical features. We model categorical features as SOS-1 sets and formulate the problem of designing poisoning attacks as a bilevel optimization problem that is nonconvex mixed-integer in the upper-level and unconstrained convex quadratic in the lower-level. We present the mathematical formulation of the problem, introduce a single-level reformulation based on the Karush-Kuhn-Tucker (KKT) conditions of the lower level, find bounds for the lower-level variables to accelerate solver performance, and propose a new algorithm to poison categorical features. Numerical experiments show that our method improves the mean squared error of all datasets compared to the previous benchmark in the literature.
☆ MOS-Attack: A Scalable Multi-objective Adversarial Attack Framework CVPR 2025
Crafting adversarial examples is crucial for evaluating and enhancing the robustness of Deep Neural Networks (DNNs), presenting a challenge equivalent to maximizing a non-differentiable 0-1 loss function. However, existing single objective methods, namely adversarial attacks focus on a surrogate loss function, do not fully harness the benefits of engaging multiple loss functions, as a result of insufficient understanding of their synergistic and conflicting nature. To overcome these limitations, we propose the Multi-Objective Set-based Attack (MOS Attack), a novel adversarial attack framework leveraging multiple loss functions and automatically uncovering their interrelations. The MOS Attack adopts a set-based multi-objective optimization strategy, enabling the incorporation of numerous loss functions without additional parameters. It also automatically mines synergistic patterns among various losses, facilitating the generation of potent adversarial attacks with fewer objectives. Extensive experiments have shown that our MOS Attack outperforms single-objective attacks. Furthermore, by harnessing the identified synergistic patterns, MOS Attack continues to show superior results with a reduced number of loss functions.
comment: Under Review of CVPR 2025
☆ Interpretable machine-learning for predicting molecular weight of PLA based on artificial bee colony optimization algorithm and adaptive neurofuzzy inference system
This article discusses the integration of the Artificial Bee Colony (ABC) algorithm with two supervised learning methods, namely Artificial Neural Networks (ANNs) and Adaptive Network-based Fuzzy Inference System (ANFIS), for feature selection from Near-Infrared (NIR) spectra for predicting the molecular weight of medical-grade Polylactic Acid (PLA). During extrusion processing of PLA, in-line NIR spectra were captured along with extrusion process and machine setting data. With a dataset comprising 63 observations and 512 input features, appropriate machine learning tools are essential for interpreting data and selecting features to improve prediction accuracy. Initially, the ABC optimization algorithm is coupled with ANN/ANFIS to forecast PLA molecular weight. The objective functions of the ABC algorithm are to minimize the root mean square error (RMSE) between experimental and predicted PLA molecular weights while also minimizing the number of input features. Results indicate that employing ABC-ANFIS yields the lowest RMSE of 282 Da and identifies four significant parameters (NIR wavenumbers 6158 cm-1, 6310 cm-1, 6349 cm-1, and melt temperature) for prediction. These findings demonstrate the effectiveness of using the ABC algorithm with ANFIS for selecting a minimal set of features to predict PLA molecular weight with high accuracy during processing
☆ Breaking Memory Limits: Gradient Wavelet Transform Enhances LLMs Training
Large language models (LLMs) have shown impressive performance across a range of natural language processing tasks. However, their vast number of parameters introduces significant memory challenges during training, particularly when using memory-intensive optimizers like Adam. Existing memory-efficient algorithms often rely on techniques such as singular value decomposition projection or weight freezing. While these approaches help alleviate memory constraints, they generally produce suboptimal results compared to full-rank updates. In this paper, we investigate the memory-efficient method beyond low-rank training, proposing a novel solution called Gradient Wavelet Transform (GWT), which applies wavelet transforms to gradients in order to significantly reduce the memory requirements for maintaining optimizer states. We demonstrate that GWT can be seamlessly integrated with memory-intensive optimizers, enabling efficient training without sacrificing performance. Through extensive experiments on both pre-training and fine-tuning tasks, we show that GWT achieves state-of-the-art performance compared with advanced memory-efficient optimizers and full-rank approaches in terms of both memory usage and training performance.
☆ A data-driven approach to discover and quantify systemic lupus erythematosus etiological heterogeneity from electronic health records
Systemic lupus erythematosus (SLE) is a complex heterogeneous disease with many manifestational facets. We propose a data-driven approach to discover probabilistic independent sources from multimodal imperfect EHR data. These sources represent exogenous variables in the data generation process causal graph that estimate latent root causes of the presence of SLE in the health record. We objectively evaluated the sources against the original variables from which they were discovered by training supervised models to discriminate SLE from negative health records using a reduced set of labelled instances. We found 19 predictive sources with high clinical validity and whose EHR signatures define independent factors of SLE heterogeneity. Using the sources as input patient data representation enables models to provide with rich explanations that better capture the clinical reasons why a particular record is (not) an SLE case. Providers may be willing to trade patient-level interpretability for discrimination especially in challenging cases.
comment: Received Runner-up Knowledge Discovery and Data Mining Innovation Award at the American Medical Informatics Association Annual Symposium 2024
☆ An Enhanced Zeroth-Order Stochastic Frank-Wolfe Framework for Constrained Finite-Sum Optimization
We propose an enhanced zeroth-order stochastic Frank-Wolfe framework to address constrained finite-sum optimization problems, a structure prevalent in large-scale machine-learning applications. Our method introduces a novel double variance reduction framework that effectively reduces the gradient approximation variance induced by zeroth-order oracles and the stochastic sampling variance from finite-sum objectives. By leveraging this framework, our algorithm achieves significant improvements in query efficiency, making it particularly well-suited for high-dimensional optimization tasks. Specifically, for convex objectives, the algorithm achieves a query complexity of O(d \sqrt{n}/\epsilon ) to find an epsilon-suboptimal solution, where d is the dimensionality and n is the number of functions in the finite-sum objective. For non-convex objectives, it achieves a query complexity of O(d^{3/2}\sqrt{n}/\epsilon^2 ) without requiring the computation ofd partial derivatives at each iteration. These complexities are the best known among zeroth-order stochastic Frank-Wolfe algorithms that avoid explicit gradient calculations. Empirical experiments on convex and non-convex machine learning tasks, including sparse logistic regression, robust classification, and adversarial attacks on deep networks, validate the computational efficiency and scalability of our approach. Our algorithm demonstrates superior performance in both convergence rate and query complexity compared to existing methods.
comment: 35 pages, 4 figures, 3 tables
☆ Lung Cancer detection using Deep Learning
In this paper we discuss lung cancer detection using hybrid model of Convolutional-Neural-Networks (CNNs) and Support-Vector-Machines-(SVMs) in order to gain early detection of tumors, benign or malignant. The work uses this hybrid model by training upon the Computed Tomography scans (CT scans) as dataset. Using deep learning for detecting lung cancer early is a cutting-edge method.
Pre-Trained Large Language Model Based Remaining Useful Life Transfer Prediction of Bearing
Accurately predicting the remaining useful life (RUL) of rotating machinery, such as bearings, is essential for ensuring equipment reliability and minimizing unexpected industrial failures. Traditional data-driven deep learning methods face challenges in practical settings due to inconsistent training and testing data distributions and limited generalization for long-term predictions.
☆ Generalizable Graph Neural Networks for Robust Power Grid Topology Control
The energy transition necessitates new congestion management methods. One such method is controlling the grid topology with machine learning (ML). This approach has gained popularity following the Learning to Run a Power Network (L2RPN) competitions. Graph neural networks (GNNs) are a class of ML models that reflect graph structure in their computation, which makes them suitable for power grid modeling. Various GNN approaches for topology control have thus been proposed. We propose the first GNN model for grid topology control that uses only GNN layers. Additionally, we identify the busbar information asymmetry problem that the popular homogeneous graph representation suffers from, and propose a heterogeneous graph representation to resolve it. We train both homogeneous and heterogeneous GNNs and fully connected neural networks (FCNN) baselines on an imitation learning task. We evaluate the models according to their classification accuracy and grid operation ability. We find that the heterogeneous GNNs perform best on in-distribution networks, followed by the FCNNs, and lastly, the homogeneous GNNs. We also find that both GNN types generalize better to out-of-distribution networks than FCNNs.
☆ Uncertainty Guarantees on Automated Precision Weeding using Conformal Prediction
Precision agriculture in general, and precision weeding in particular, have greatly benefited from the major advancements in deep learning and computer vision. A large variety of commercial robotic solutions are already available and deployed. However, the adoption by farmers of such solutions is still low for many reasons, an important one being the lack of trust in these systems. This is in great part due to the opaqueness and complexity of deep neural networks and the manufacturers' inability to provide valid guarantees on their performance. Conformal prediction, a well-established methodology in the machine learning community, is an efficient and reliable strategy for providing trustworthy guarantees on the predictions of any black-box model under very minimal constraints. Bridging the gap between the safe machine learning and precision agriculture communities, this article showcases conformal prediction in action on the task of precision weeding through deep learning-based image classification. After a detailed presentation of the conformal prediction methodology and the development of a precision spraying pipeline based on a ''conformalized'' neural network and well-defined spraying decision rules, the article evaluates this pipeline on two real-world scenarios: one under in-distribution conditions, the other reflecting a near out-of-distribution setting. The results show that we are able to provide formal, i.e. certifiable, guarantees on spraying at least 90% of the weeds.
☆ Knowledge Distillation and Enhanced Subdomain Adaptation Using Graph Convolutional Network for Resource-Constrained Bearing Fault Diagnosis
Bearing fault diagnosis under varying working conditions faces challenges, including a lack of labeled data, distribution discrepancies, and resource constraints. To address these issues, we propose a progressive knowledge distillation framework that transfers knowledge from a complex teacher model, utilizing a Graph Convolutional Network (GCN) with Autoregressive moving average (ARMA) filters, to a compact and efficient student model. To mitigate distribution discrepancies and labeling uncertainty, we introduce Enhanced Local Maximum Mean Squared Discrepancy (ELMMSD), which leverages mean and variance statistics in the Reproducing Kernel Hilbert Space (RKHS) and incorporates a priori probability distributions between labels. This approach increases the distance between clustering centers, bridges subdomain gaps, and enhances subdomain alignment reliability. Experimental results on benchmark datasets (CWRU and JNU) demonstrate that the proposed method achieves superior diagnostic accuracy while significantly reducing computational costs. Comprehensive ablation studies validate the effectiveness of each component, highlighting the robustness and adaptability of the approach across diverse working conditions.
☆ Anomalous Agreement: How to find the Ideal Number of Anomaly Classes in Correlated, Multivariate Time Series Data AAAI
Detecting and classifying abnormal system states is critical for condition monitoring, but supervised methods often fall short due to the rarity of anomalies and the lack of labeled data. Therefore, clustering is often used to group similar abnormal behavior. However, evaluating cluster quality without ground truth is challenging, as existing measures such as the Silhouette Score (SSC) only evaluate the cohesion and separation of clusters and ignore possible prior knowledge about the data. To address this challenge, we introduce the Synchronized Anomaly Agreement Index (SAAI), which exploits the synchronicity of anomalies across multivariate time series to assess cluster quality. We demonstrate the effectiveness of SAAI by showing that maximizing SAAI improves accuracy on the task of finding the true number of anomaly classes K in correlated time series by 0.23 compared to SSC and by 0.32 compared to X-Means. We also show that clusters obtained by maximizing SAAI are easier to interpret compared to SSC.
comment: Acccepted at AAAI Workshop on AI for Time Series Analysis (AI4TS) 2025
☆ AlphaNet: Scaling Up Local Frame-based Atomistic Foundation Model
We present AlphaNet, a local frame-based equivariant model designed to achieve both accurate and efficient simulations for atomistic systems. Recently, machine learning force fields (MLFFs) have gained prominence in molecular dynamics simulations due to their advantageous efficiency-accuracy balance compared to classical force fields and quantum mechanical calculations, alongside their transferability across various systems. Despite the advancements in improving model accuracy, the efficiency and scalability of MLFFs remain significant obstacles in practical applications. AlphaNet enhances computational efficiency and accuracy by leveraging the local geometric structures of atomic environments through the construction of equivariant local frames and learnable frame transitions. We substantiate the efficacy of AlphaNet across diverse datasets, including defected graphene, formate decomposition, zeolites, and surface reactions. AlphaNet consistently surpasses well-established models, such as NequIP and DeepPot, in terms of both energy and force prediction accuracy. Notably, AlphaNet offers one of the best trade-offs between computational efficiency and accuracy among existing models. Moreover, AlphaNet exhibits scalability across a broad spectrum of system and dataset sizes, affirming its versatility.
comment: 14 pages, 5 figures
☆ TIMRL: A Novel Meta-Reinforcement Learning Framework for Non-Stationary and Multi-Task Environments
In recent years, meta-reinforcement learning (meta-RL) algorithm has been proposed to improve sample efficiency in the field of decision-making and control, enabling agents to learn new knowledge from a small number of samples. However, most research uses the Gaussian distribution to extract task representation, which is poorly adapted to tasks that change in non-stationary environment. To address this problem, we propose a novel meta-reinforcement learning method by leveraging Gaussian mixture model and the transformer network to construct task inference model. The Gaussian mixture model is utilized to extend the task representation and conduct explicit encoding of tasks. Specifically, the classification of tasks is encoded through transformer network to determine the Gaussian component corresponding to the task. By leveraging task labels, the transformer network is trained using supervised learning. We validate our method on MuJoCo benchmarks with non-stationary and multi-task environments. Experimental results demonstrate that the proposed method dramatically improves sample efficiency and accurately recognizes the classification of the tasks, while performing excellently in the environment.
☆ A User's Guide to $\texttt{KSig}$: GPU-Accelerated Computation of the Signature Kernel
The signature kernel is a positive definite kernel for sequential and temporal data that has become increasingly popular in machine learning applications due to powerful theoretical guarantees, strong empirical performance, and recently introduced various scalable variations. In this chapter, we give a short introduction to $\texttt{KSig}$, a $\texttt{Scikit-Learn}$ compatible Python package that implements various GPU-accelerated algorithms for computing signature kernels, and performing downstream learning tasks. We also introduce a new algorithm based on tensor sketches which gives strong performance compared to existing algorithms. The package is available at $\href{https://github.com/tgcsaba/ksig}{\texttt{https://github.com/tgcsaba/ksig}}$.
☆ LLM360 K2: Scaling Up 360-Open-Source Large Language Models
We detail the training of the LLM360 K2-65B model, scaling up our 360-degree OPEN SOURCE approach to the largest and most powerful models under project LLM360. While open-source LLMs continue to advance, the answer to "How are the largest LLMs trained?" remains unclear within the community. The implementation details for such high-capacity models are often protected due to business considerations associated with their high cost. This lack of transparency prevents LLM researchers from leveraging valuable insights from prior experience, e.g., "What are the best practices for addressing loss spikes?" The LLM360 K2 project addresses this gap by providing full transparency and access to resources accumulated during the training of LLMs at the largest scale. This report highlights key elements of the K2 project, including our first model, K2 DIAMOND, a 65 billion-parameter LLM that surpasses LLaMA-65B and rivals LLaMA2-70B, while requiring fewer FLOPs and tokens. We detail the implementation steps and present a longitudinal analysis of K2 DIAMOND's capabilities throughout its training process. We also outline ongoing projects such as TXT360, setting the stage for future models in the series. By offering previously unavailable resources, the K2 project also resonates with the 360-degree OPEN SOURCE principles of transparency, reproducibility, and accessibility, which we believe are vital in the era of resource-intensive AI research.
☆ Inferring Interpretable Models of Fragmentation Functions using Symbolic Regression
Machine learning is rapidly making its path into natural sciences, including high-energy physics. We present the first study that infers, directly from experimental data, a functional form of fragmentation functions. The latter represent a key ingredient to describe physical observables measured in high-energy physics processes that involve hadron production, and predict their values at different energy. Fragmentation functions can not be calculated in theory and have to be determined instead from data. Traditional approaches rely on global fits of experimental data using a pre-assumed functional form inspired from phenomenological models to learn its parameters. This novel approach uses a ML technique, namely symbolic regression, to learn an analytical model from measured charged hadron multiplicities. The function learned by symbolic regression resembles the Lund string function and describes the data well, thus representing a potential candidate for use in global FFs fits. This study represents an approach to follow in such QCD-related phenomenology studies and more generally in sciences.
☆ D3MES: Diffusion Transformer with multihead equivariant self-attention for 3D molecule generation
Understanding and predicting the diverse conformational states of molecules is crucial for advancing fields such as chemistry, material science, and drug development. Despite significant progress in generative models, accurately generating complex and biologically or material-relevant molecular structures remains a major challenge. In this work, we introduce a diffusion model for three-dimensional (3D) molecule generation that combines a classifiable diffusion model, Diffusion Transformer, with multihead equivariant self-attention. This method addresses two key challenges: correctly attaching hydrogen atoms in generated molecules through learning representations of molecules after hydrogen atoms are removed; and overcoming the limitations of existing models that cannot generate molecules across multiple classes simultaneously. The experimental results demonstrate that our model not only achieves state-of-the-art performance across several key metrics but also exhibits robustness and versatility, making it highly suitable for early-stage large-scale generation processes in molecular design, followed by validation and further screening to obtain molecules with specific properties.
☆ SFC-GAN: A Generative Adversarial Network for Brain Functional and Structural Connectome Translation
Modern brain imaging technologies have enabled the detailed reconstruction of human brain connectomes, capturing structural connectivity (SC) from diffusion MRI and functional connectivity (FC) from functional MRI. Understanding the intricate relationships between SC and FC is vital for gaining deeper insights into the brain's functional and organizational mechanisms. However, obtaining both SC and FC modalities simultaneously remains challenging, hindering comprehensive analyses. Existing deep generative models typically focus on synthesizing a single modality or unidirectional translation between FC and SC, thereby missing the potential benefits of bi-directional translation, especially in scenarios where only one connectome is available. Therefore, we propose Structural-Functional Connectivity GAN (SFC-GAN), a novel framework for bidirectional translation between SC and FC. This approach leverages the CycleGAN architecture, incorporating convolutional layers to effectively capture the spatial structures of brain connectomes. To preserve the topological integrity of these connectomes, we employ a structure-preserving loss that guides the model in capturing both global and local connectome patterns while maintaining symmetry. Our framework demonstrates superior performance in translating between SC and FC, outperforming baseline models in similarity and graph property evaluations compared to ground truth data, each translated modality can be effectively utilized for downstream classification.
comment: 5 pages, 2 figures
☆ Differentially Private Kernelized Contextual Bandits
We consider the problem of contextual kernel bandits with stochastic contexts, where the underlying reward function belongs to a known Reproducing Kernel Hilbert Space (RKHS). We study this problem under the additional constraint of joint differential privacy, where the agents needs to ensure that the sequence of query points is differentially private with respect to both the sequence of contexts and rewards. We propose a novel algorithm that improves upon the state of the art and achieves an error rate of $\mathcal{O}\left(\sqrt{\frac{\gamma_T}{T}} + \frac{\gamma_T}{T \varepsilon}\right)$ after $T$ queries for a large class of kernel families, where $\gamma_T$ represents the effective dimensionality of the kernel and $\varepsilon > 0$ is the privacy parameter. Our results are based on a novel estimator for the reward function that simultaneously enjoys high utility along with a low-sensitivity to observed rewards and contexts, which is crucial to obtain an order optimal learning performance with improved dependence on the privacy parameter.
☆ ACCon: Angle-Compensated Contrastive Regularizer for Deep Regression AAAI-2025
In deep regression, capturing the relationship among continuous labels in feature space is a fundamental challenge that has attracted increasing interest. Addressing this issue can prevent models from converging to suboptimal solutions across various regression tasks, leading to improved performance, especially for imbalanced regression and under limited sample sizes. However, existing approaches often rely on order-aware representation learning or distance-based weighting. In this paper, we hypothesize a linear negative correlation between label distances and representation similarities in regression tasks. To implement this, we propose an angle-compensated contrastive regularizer for deep regression, which adjusts the cosine distance between anchor and negative samples within the contrastive learning framework. Our method offers a plug-and-play compatible solution that extends most existing contrastive learning methods for regression tasks. Extensive experiments and theoretical analysis demonstrate that our proposed angle-compensated contrastive regularizer not only achieves competitive regression performance but also excels in data efficiency and effectiveness on imbalanced datasets.
comment: Accept by AAAI-2025 (The 39th Annual AAAI Conference on Artificial Intelligence)
☆ Protego: Detecting Adversarial Examples for Vision Transformers via Intrinsic Capabilities
Transformer models have excelled in natural language tasks, prompting the vision community to explore their implementation in computer vision problems. However, these models are still influenced by adversarial examples. In this paper, we investigate the attack capabilities of six common adversarial attacks on three pretrained ViT models to reveal the vulnerability of ViT models. To understand and analyse the bias in neural network decisions when the input is adversarial, we use two visualisation techniques that are attention rollout and grad attention rollout. To prevent ViT models from adversarial attack, we propose Protego, a detection framework that leverages the transformer intrinsic capabilities to detection adversarial examples of ViT models. Nonetheless, this is challenging due to a diversity of attack strategies that may be adopted by adversaries. Inspired by the attention mechanism, we know that the token of prediction contains all the information from the input sample. Additionally, the attention region for adversarial examples differs from that of normal examples. Given these points, we can train a detector that achieves superior performance than existing detection methods to identify adversarial examples. Our experiments have demonstrated the high effectiveness of our detection method. For these six adversarial attack methods, our detector's AUC scores all exceed 0.95. Protego may advance investigations in metaverse security.
comment: Accepted by IEEE MetaCom 2024
☆ Explore the Use of Time Series Foundation Model for Car-Following Behavior Analysis
Modeling car-following behavior is essential for traffic simulation, analyzing driving patterns, and understanding complex traffic flows with varying levels of autonomous vehicles. Traditional models like the Safe Distance Model and Intelligent Driver Model (IDM) require precise parameter calibration and often lack generality due to simplified assumptions about driver behavior. While machine learning and deep learning methods capture complex patterns, they require large labeled datasets. Foundation models provide a more efficient alternative. Pre-trained on vast, diverse time series datasets, they can be applied directly to various tasks without the need for extensive re-training. These models generalize well across domains, and with minimal fine-tuning, they can be adapted to specific tasks like car-following behavior prediction. In this paper, we apply Chronos, a state-of-the-art public time series foundation model, to analyze car-following behavior using the Open ACC dataset. Without fine-tuning, Chronos outperforms traditional models like IDM and Exponential smoothing with trend and seasonality (ETS), and achieves similar results to deep learning models such as DeepAR and TFT, with an RMSE of 0.60. After fine-tuning, Chronos reduces the error to an RMSE of 0.53, representing a 33.75% improvement over IDM and a 12-37% reduction compared to machine learning models like ETS and deep learning models including DeepAR, WaveNet, and TFT. This demonstrates the potential of foundation models to significantly advance transportation research, offering a scalable, adaptable, and highly accurate approach to predicting and simulating car-following behaviors.
☆ Detection of AI Deepfake and Fraud in Online Payments Using GAN-Based Models
This study explores the use of Generative Adversarial Networks (GANs) to detect AI deepfakes and fraudulent activities in online payment systems. With the growing prevalence of deepfake technology, which can manipulate facial features in images and videos, the potential for fraud in online transactions has escalated. Traditional security systems struggle to identify these sophisticated forms of fraud. This research proposes a novel GAN-based model that enhances online payment security by identifying subtle manipulations in payment images. The model is trained on a dataset consisting of real-world online payment images and deepfake images generated using advanced GAN architectures, such as StyleGAN and DeepFake. The results demonstrate that the proposed model can accurately distinguish between legitimate transactions and deepfakes, achieving a high detection rate above 95%. This approach significantly improves the robustness of payment systems against AI-driven fraud. The paper contributes to the growing field of digital security, offering insights into the application of GANs for fraud detection in financial services. Keywords- Payment Security, Image Recognition, Generative Adversarial Networks, AI Deepfake, Fraudulent Activities
comment: The paper will be published and indexed by IEEE at 2025 8th International Conference on Advanced Algorithms and Control Engineering (ICAACE 2025)
☆ PRKAN: Parameter-Reduced Kolmogorov-Arnold Networks
Kolmogorov-Arnold Networks (KANs) represent an innovation in neural network architectures, offering a compelling alternative to Multi-Layer Perceptrons (MLPs) in models such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Transformers. By advancing network design, KANs are driving groundbreaking research and enabling transformative applications across various scientific domains involving neural networks. However, existing KANs often require significantly more parameters in their network layers compared to MLPs. To address this limitation, this paper introduces PRKANs (\textbf{P}arameter-\textbf{R}educed \textbf{K}olmogorov-\textbf{A}rnold \textbf{N}etworks), which employ several methods to reduce the parameter count in KAN layers, making them comparable to MLP layers. Experimental results on the MNIST and Fashion-MNIST datasets demonstrate that PRKANs with attention mechanisms outperform several existing KANs and rival the performance of MLPs, albeit with slightly longer training times. Furthermore, the study highlights the advantages of Gaussian Radial Basis Functions (GRBFs) and layer normalization in KAN designs. The repository for this work is available at: \url{https://github.com/hoangthangta/All-KAN}.
comment: 23 pages
☆ Erasing Noise in Signal Detection with Diffusion Model: From Theory to Application
In this paper, a signal detection method based on the denoise diffusion model (DM) is proposed, which outperforms the maximum likelihood (ML) estimation method that has long been regarded as the optimal signal detection technique. Theoretically, a novel mathematical theory for intelligent signal detection based on stochastic differential equations (SDEs) is established in this paper, demonstrating the effectiveness of DM in reducing the additive white Gaussian noise in received signals. Moreover, a mathematical relationship between the signal-to-noise ratio (SNR) and the timestep in DM is established, revealing that for any given SNR, a corresponding optimal timestep can be identified. Furthermore, to address potential issues with out-of-distribution inputs in the DM, we employ a mathematical scaling technique that allows the trained DM to handle signal detection across a wide range of SNRs without any fine-tuning. Building on the above theoretical foundation, we propose a DM-based signal detection method, with the diffusion transformer (DiT) serving as the backbone neural network, whose computational complexity of this method is $\mathcal{O}(n^2)$. Simulation results demonstrate that, for BPSK and QAM modulation schemes, the DM-based method achieves a significantly lower symbol error rate (SER) compared to ML estimation, while maintaining a much lower computational complexity.
☆ Improved Regret Bounds for Online Fair Division with Bandit Learning
We study online fair division when there are a finite number of item types and the player values for the items are drawn randomly from distributions with unknown means. In this setting, a sequence of indivisible items arrives according to a random online process, and each item must be allocated to a single player. The goal is to maximize expected social welfare while maintaining that the allocation satisfies proportionality in expectation. When player values are normalized, we show that it is possible to with high probability guarantee proportionality constraint satisfaction and achieve $\tilde{O}(\sqrt{T})$ regret. To achieve this result, we present an upper confidence bound (UCB) algorithm that uses two rounds of linear optimization. This algorithm highlights fundamental aspects of proportionality constraints that allow for a UCB algorithm despite the presence of many (potentially tight) constraints. This result improves upon the previous best regret rate of $\tilde{O}(T^{2/3})$.
☆ Neural Probabilistic Circuits: Enabling Compositional and Interpretable Predictions through Logical Reasoning
End-to-end deep neural networks have achieved remarkable success across various domains but are often criticized for their lack of interpretability. While post hoc explanation methods attempt to address this issue, they often fail to accurately represent these black-box models, resulting in misleading or incomplete explanations. To overcome these challenges, we propose an inherently transparent model architecture called Neural Probabilistic Circuits (NPCs), which enable compositional and interpretable predictions through logical reasoning. In particular, an NPC consists of two modules: an attribute recognition model, which predicts probabilities for various attributes, and a task predictor built on a probabilistic circuit, which enables logical reasoning over recognized attributes to make class predictions. To train NPCs, we introduce a three-stage training algorithm comprising attribute recognition, circuit construction, and joint optimization. Moreover, we theoretically demonstrate that an NPC's error is upper-bounded by a linear combination of the errors from its modules. To further demonstrate the interpretability of NPC, we provide both the most probable explanations and the counterfactual explanations. Empirical results on four benchmark datasets show that NPCs strike a balance between interpretability and performance, achieving results competitive even with those of end-to-end black-box models while providing enhanced interpretability.
☆ AlgoRxplorers | Precision in Mutation -- Enhancing Drug Design with Advanced Protein Stability Prediction Tools
Predicting the impact of single-point amino acid mutations on protein stability is essential for understanding disease mechanisms and advancing drug development. Protein stability, quantified by changes in Gibbs free energy ($\Delta\Delta G$), is influenced by these mutations. However, the scarcity of data and the complexity of model interpretation pose challenges in accurately predicting stability changes. This study proposes the application of deep neural networks, leveraging transfer learning and fusing complementary information from different models, to create a feature-rich representation of the protein stability landscape. We developed four models, with our third model, ThermoMPNN+, demonstrating the best performance in predicting $\Delta\Delta G$ values. This approach, which integrates diverse feature sets and embeddings through latent transfusion techniques, aims to refine $\Delta\Delta G$ predictions and contribute to a deeper understanding of protein dynamics, potentially leading to advancements in disease research and drug discovery.
☆ Global Search for Optimal Low Thrust Spacecraft Trajectories using Diffusion Models and the Indirect Method
Long time-duration low-thrust nonlinear optimal spacecraft trajectory global search is a computationally and time expensive problem characterized by clustering patterns in locally optimal solutions. During preliminary mission design, mission parameters are subject to frequent changes, necessitating that trajectory designers efficiently generate high-quality control solutions for these new scenarios. Generative machine learning models can be trained to learn how the solution structure varies with respect to a conditional parameter, thereby accelerating the global search for missions with updated parameters. In this work, state-of-the-art diffusion models are integrated with the indirect approach for trajectory optimization within a global search framework. This framework is tested on two low-thrust transfers of different complexity in the circular restricted three-body problem. By generating and analyzing a training data set, we develop mathematical relations and techniques to understand the complex structures in the costate domain of locally optimal solutions for these problems. A diffusion model is trained on this data and successfully accelerates the global search for both problems. The model predicts how the costate solution structure changes, based on the maximum spacecraft thrust magnitude. Warm-starting a numerical solver with diffusion model samples for the costates at the initial time increases the number of solutions generated per minute for problems with unseen thrust magnitudes by one to two orders of magnitude in comparison to samples from a uniform distribution and from an adjoint control transformation.
☆ Likelihood Training of Cascaded Diffusion Models via Hierarchical Volume-preserving Maps ICLR 2024
Cascaded models are multi-scale generative models with a marked capacity for producing perceptually impressive samples at high resolutions. In this work, we show that they can also be excellent likelihood models, so long as we overcome a fundamental difficulty with probabilistic multi-scale models: the intractability of the likelihood function. Chiefly, in cascaded models each intermediary scale introduces extraneous variables that cannot be tractably marginalized out for likelihood evaluation. This issue vanishes by modeling the diffusion process on latent spaces induced by a class of transformations we call hierarchical volume-preserving maps, which decompose spatially structured data in a hierarchical fashion without introducing local distortions in the latent space. We demonstrate that two such maps are well-known in the literature for multiscale modeling: Laplacian pyramids and wavelet transforms. Not only do such reparameterizations allow the likelihood function to be directly expressed as a joint likelihood over the scales, we show that the Laplacian pyramid and wavelet transform also produces significant improvements to the state-of-the-art on a selection of benchmarks in likelihood modeling, including density estimation, lossless compression, and out-of-distribution detection. Investigating the theoretical basis of our empirical gains we uncover deep connections to score matching under the Earth Mover's Distance (EMD), which is a well-known surrogate for perceptual similarity. Code can be found at \href{https://github.com/lihenryhfl/pcdm}{this https url}.
comment: Spotlight at ICLR 2024
☆ Motion Tracks: A Unified Representation for Human-Robot Transfer in Few-Shot Imitation Learning
Teaching robots to autonomously complete everyday tasks remains a challenge. Imitation Learning (IL) is a powerful approach that imbues robots with skills via demonstrations, but is limited by the labor-intensive process of collecting teleoperated robot data. Human videos offer a scalable alternative, but it remains difficult to directly train IL policies from them due to the lack of robot action labels. To address this, we propose to represent actions as short-horizon 2D trajectories on an image. These actions, or motion tracks, capture the predicted direction of motion for either human hands or robot end-effectors. We instantiate an IL policy called Motion Track Policy (MT-pi) which receives image observations and outputs motion tracks as actions. By leveraging this unified, cross-embodiment action space, MT-pi completes tasks with high success given just minutes of human video and limited additional robot demonstrations. At test time, we predict motion tracks from two camera views, recovering 6DoF trajectories via multi-view synthesis. MT-pi achieves an average success rate of 86.5% across 4 real-world tasks, outperforming state-of-the-art IL baselines which do not leverage human data or our action space by 40%, and generalizes to scenarios seen only in human videos. Code and videos are available on our website https://portal-cornell.github.io/motion_track_policy/.
☆ Combining LLM decision and RL action selection to improve RL policy for adaptive interventions
Reinforcement learning (RL) is increasingly being used in the healthcare domain, particularly for the development of personalized health adaptive interventions. Inspired by the success of Large Language Models (LLMs), we are interested in using LLMs to update the RL policy in real time, with the goal of accelerating personalization. We use the text-based user preference to influence the action selection on the fly, in order to immediately incorporate the user preference. We use the term "user preference" as a broad term to refer to a user personal preference, constraint, health status, or a statement expressing like or dislike, etc. Our novel approach is a hybrid method that combines the LLM response and the RL action selection to improve the RL policy. Given an LLM prompt that incorporates the user preference, the LLM acts as a filter in the typical RL action selection. We investigate different prompting strategies and action selection strategies. To evaluate our approach, we implement a simulation environment that generates the text-based user preferences and models the constraints that impact behavioral dynamics. We show that our approach is able to take into account the text-based user preferences, while improving the RL policy, thus improving personalization in adaptive intervention.
♻ ☆ SecAlign: Defending Against Prompt Injection with Preference Optimization
Large language models (LLMs) are becoming increasingly prevalent in modern software systems, interfacing between the user and the Internet to assist with tasks that require advanced language understanding. To accomplish these tasks, the LLM often uses external data sources such as user documents, web retrieval, results from API calls, etc. This opens up new avenues for attackers to manipulate the LLM via prompt injection. Adversarial prompts can be injected into external data sources to override the system's intended instruction and instead execute a malicious instruction. To mitigate this vulnerability, we propose a new defense called SecAlign based on the technique of preference optimization. Our defense first constructs a preference dataset with prompt-injected inputs, secure outputs (ones that respond to the legitimate instruction), and insecure outputs (ones that respond to the injection). We then perform preference optimization on this dataset to teach the LLM to prefer the secure output over the insecure one. This provides the first known method that reduces the success rates of various prompt injections to around 0%, even against attacks much more sophisticated than ones seen during training. This indicates our defense generalizes well against unknown and yet-to-come attacks. Also, our defended models are still practical with similar utility to the one before our defensive training. Our code is at https://github.com/facebookresearch/SecAlign
comment: Key words: prompt injection defense, LLM security, LLM-integrated applications
♻ ☆ Geometric Scattering on Measure Spaces
The scattering transform is a multilayered, wavelet-based transform initially introduced as a model of convolutional neural networks (CNNs) that has played a foundational role in our understanding of these networks' stability and invariance properties. Subsequently, there has been widespread interest in extending the success of CNNs to data sets with non-Euclidean structure, such as graphs and manifolds, leading to the emerging field of geometric deep learning. In order to improve our understanding of the architectures used in this new field, several papers have proposed generalizations of the scattering transform for non-Euclidean data structures such as undirected graphs and compact Riemannian manifolds without boundary. In this paper, we introduce a general, unified model for geometric scattering on measure spaces. Our proposed framework includes previous work on geometric scattering as special cases but also applies to more general settings such as directed graphs, signed graphs, and manifolds with boundary. We propose a new criterion that identifies to which groups a useful representation should be invariant and show that this criterion is sufficient to guarantee that the scattering transform has desirable stability and invariance properties. Additionally, we consider finite measure spaces that are obtained from randomly sampling an unknown manifold. We propose two methods for constructing a data-driven graph on which the associated graph scattering transform approximates the scattering transform on the underlying manifold. Moreover, we use a diffusion-maps based approach to prove quantitative estimates on the rate of convergence of one of these approximations as the number of sample points tends to infinity. Lastly, we showcase the utility of our method on spherical images, directed graphs, and on high-dimensional single-cell data.
♻ ☆ Few-Shot Task Learning through Inverse Generative Modeling
Learning the intents of an agent, defined by its goals or motion style, is often extremely challenging from just a few examples. We refer to this problem as task concept learning and present our approach, Few-Shot Task Learning through Inverse Generative Modeling (FTL-IGM), which learns new task concepts by leveraging invertible neural generative models. The core idea is to pretrain a generative model on a set of basic concepts and their demonstrations. Then, given a few demonstrations of a new concept (such as a new goal or a new action), our method learns the underlying concepts through backpropagation without updating the model weights, thanks to the invertibility of the generative model. We evaluate our method in five domains -- object rearrangement, goal-oriented navigation, motion caption of human actions, autonomous driving, and real-world table-top manipulation. Our experimental results demonstrate that via the pretrained generative model, we successfully learn novel concepts and generate agent plans or motion corresponding to these concepts in (1) unseen environments and (2) in composition with training concepts.
comment: Added acknowledgment
♻ ☆ Improving the Performance of Echo State Networks Through State Feedback
Reservoir computing, using nonlinear dynamical systems, offers a cost-effective alternative to neural networks for complex tasks involving processing of sequential data, time series modeling, and system identification. Echo state networks (ESNs), a type of reservoir computer, mirror neural networks but simplify training. They apply fixed, random linear transformations to the internal state, followed by nonlinear changes. This process, guided by input signals and linear regression, adapts the system to match target characteristics, reducing computational demands. A potential drawback of ESNs is that the fixed reservoir may not offer the complexity needed for specific problems. While directly altering (training) the internal ESN would reintroduce the computational burden, an indirect modification can be achieved by redirecting some output as input. This feedback can influence the internal reservoir state, yielding ESNs with enhanced complexity suitable for broader challenges. In this paper, we demonstrate that by feeding some component of the reservoir state back into the network through the input, we can drastically improve upon the performance of a given ESN. We rigorously prove that, for any given ESN, feedback will almost always improve the accuracy of the output. For a set of three tasks, each representing different problem classes, we find that with feedback the average error measures are reduced by $30\%-60\%$. Remarkably, feedback provides at least an equivalent performance boost to doubling the initial number of computational nodes, a computationally expensive and technologically challenging alternative. These results demonstrate the broad applicability and substantial usefulness of this feedback scheme.
comment: 36 pages, 6 figures
♻ ☆ Quilt-1M: One Million Image-Text Pairs for Histopathology
Recent accelerations in multi-modal applications have been made possible with the plethora of image and text data available online. However, the scarcity of analogous data in the medical field, specifically in histopathology, has slowed comparable progress. To enable similar representation learning for histopathology, we turn to YouTube, an untapped resource of videos, offering $1,087$ hours of valuable educational histopathology videos from expert clinicians. From YouTube, we curate QUILT: a large-scale vision-language dataset consisting of $802, 144$ image and text pairs. QUILT was automatically curated using a mixture of models, including large language models, handcrafted algorithms, human knowledge databases, and automatic speech recognition. In comparison, the most comprehensive datasets curated for histopathology amass only around $200$K samples. We combine QUILT with datasets from other sources, including Twitter, research papers, and the internet in general, to create an even larger dataset: QUILT-1M, with $1$M paired image-text samples, marking it as the largest vision-language histopathology dataset to date. We demonstrate the value of QUILT-1M by fine-tuning a pre-trained CLIP model. Our model outperforms state-of-the-art models on both zero-shot and linear probing tasks for classifying new histopathology images across $13$ diverse patch-level datasets of $8$ different sub-pathologies and cross-modal retrieval tasks.
♻ ☆ Higher-Order Topological Directionality and Directed Simplicial Neural Networks
Topological Deep Learning (TDL) has emerged as a paradigm to process and learn from signals defined on higher-order combinatorial topological spaces, such as simplicial or cell complexes. Although many complex systems have an asymmetric relational structure, most TDL models forcibly symmetrize these relationships. In this paper, we first introduce a novel notion of higher-order directionality and we then design Directed Simplicial Neural Networks (Dir-SNNs) based on it. Dir-SNNs are message-passing networks operating on directed simplicial complexes able to leverage directed and possibly asymmetric interactions among the simplices. To our knowledge, this is the first TDL model using a notion of higher-order directionality. We theoretically and empirically prove that Dir-SNNs are more expressive than their directed graph counterpart in distinguishing isomorphic directed graphs. Experiments on a synthetic source localization task demonstrate that Dir-SNNs outperform undirected SNNs when the underlying complex is directed, and perform comparably when the underlying complex is undirected.
comment: 7 pages, 8 figures, 1 table
♻ ☆ Enhance Eye Disease Detection using Learnable Probabilistic Discrete Latents in Machine Learning Architectures
Ocular diseases, including diabetic retinopathy and glaucoma, present a significant public health challenge due to their high prevalence and potential for causing vision impairment. Early and accurate diagnosis is crucial for effective treatment and management. In recent years, deep learning models have emerged as powerful tools for analysing medical images, such as retina imaging. However, challenges persist in model relibability and uncertainty estimation, which are critical for clinical decision-making. This study leverages the probabilistic framework of Generative Flow Networks (GFlowNets) to learn the posterior distribution over latent discrete dropout masks for the classification and analysis of ocular diseases using fundus images. We develop a robust and generalizable method that utilizes GFlowOut integrated with ResNet18 and ViT models as the backbone in identifying various ocular conditions. This study employs a unique set of dropout masks - none, random, bottomup, and topdown - to enhance model performance in analyzing these fundus images. Our results demonstrate that our learnable probablistic latents significantly improves accuracy, outperforming the traditional dropout approach. We utilize a gradient map calculation method, Grad-CAM, to assess model explainability, observing that the model accurately focuses on critical image regions for predictions. The integration of GFlowOut in neural networks presents a promising advancement in the automated diagnosis of ocular diseases, with implications for improving clinical workflows and patient outcomes.
♻ ☆ Path Loss Prediction Using Deep Learning
Radio deployments and spectrum planning benefit from path loss predictions. Obstructions along a communications link are often considered implicitly or through derived metrics such as representative clutter height or total obstruction depth. In this paper, we propose a path-specific path loss prediction method that uses convolutional neural networks to automatically perform feature extraction from high-resolution obstruction height maps. Our methods result in low prediction error in a variety of environments without requiring derived metrics.
comment: 5 pages, 3 figures, 4 tables
♻ ☆ FlashRNN: Optimizing Traditional RNNs on Modern Hardware
While Transformers and other sequence-parallelizable neural network architectures seem like the current state of the art in sequence modeling, they specifically lack state-tracking capabilities. These are important for time-series tasks and logical reasoning. Traditional RNNs like LSTMs and GRUs, as well as modern variants like sLSTM do have these capabilities at the cost of strictly sequential processing. While this is often seen as a strong limitation, we show how fast these networks can get with our hardware-optimization FlashRNN in Triton and CUDA, optimizing kernels to the register level on modern GPUs. We extend traditional RNNs with a parallelization variant that processes multiple RNNs of smaller hidden state in parallel, similar to the head-wise processing in Transformers. To enable flexibility on different GPU variants, we introduce a new optimization framework for hardware-internal cache sizes, memory and compute handling. It models the hardware in a setting using polyhedral-like constraints, including the notion of divisibility. This speeds up the solution process in our ConstrINT library for general integer constraint satisfaction problems (integer CSPs). We show that our kernels can achieve 50x speed-ups over a vanilla PyTorch implementation and allow 40x larger hidden sizes compared to our Triton implementation. Our open-source kernels and the optimization library are released here to boost research in the direction of state-tracking enabled RNNs and sequence modeling: \url{https://github.com/NX-AI/flashrnn}
♻ ☆ Hybrid Top-Down Global Causal Discovery with Local Search for Linear and Nonlinear Additive Noise Models NeurIPS 2024
Learning the unique directed acyclic graph corresponding to an unknown causal model is a challenging task. Methods based on functional causal models can identify a unique graph, but either suffer from the curse of dimensionality or impose strong parametric assumptions. To address these challenges, we propose a novel hybrid approach for global causal discovery in observational data that leverages local causal substructures. We first present a topological sorting algorithm that leverages ancestral relationships in linear structural causal models to establish a compact top-down hierarchical ordering, encoding more causal information than linear orderings produced by existing methods. We demonstrate that this approach generalizes to nonlinear settings with arbitrary noise. We then introduce a nonparametric constraint-based algorithm that prunes spurious edges by searching for local conditioning sets, achieving greater accuracy than current methods. We provide theoretical guarantees for correctness and worst-case polynomial time complexities, with empirical validation on synthetic data.
comment: To appear at the Thirty-Eighth Annual Conference on Neural Information Processing Systems (NeurIPS 2024)
♻ ☆ A Unified Approach to Extract Interpretable Rules from Tree Ensembles via Integer Programming
Tree ensembles are very popular machine learning models, known for their effectiveness in supervised classification and regression tasks. Their performance derives from aggregating predictions of multiple decision trees, which are renowned for their interpretability properties. However, tree ensemble models do not reliably exhibit interpretable output. Our work aims to extract an optimized list of rules from a trained tree ensemble, providing the user with a condensed, interpretable model that retains most of the predictive power of the full model. Our approach consists of solving a set partitioning problem formulated through Integer Programming. The proposed method works with either tabular or time series data, for both classification and regression tasks, and its flexible formulation can include any arbitrary loss or regularization functions. Our extensive computational experiments offer statistically significant evidence that our method is competitive with other rule extraction methods in terms of predictive performance and fidelity towards the tree ensemble. Moreover, we empirically show that the proposed method effectively extracts interpretable rules from tree ensemble that are designed for time series data.
comment: - Improved overall manuscript flow and clearness - Added related work on explanation fidelity - Added computational results on fidelity - Fixed some flaws on data inference - Optimization problem with weighted objectives - Added appendix containing qualitative examples - New computational results
♻ ☆ Steering Large Language Models using Conceptors: Improving Addition-Based Activation Engineering NeurIPS 2024
Large language models have transformed AI, yet reliably controlling their outputs remains a challenge. This paper explores activation engineering, where outputs of pre-trained LLMs are controlled by manipulating their activations at inference time. Unlike traditional methods using a single steering vector, we introduce conceptors - mathematical constructs that represent sets of activation vectors as ellipsoidal regions. Conceptors act as soft projection matrices and offer more precise control over complex activation patterns. Our experiments demonstrate that conceptors outperform traditional methods across multiple steering tasks. We further use Boolean operations on conceptors for combined steering goals that empirically outperform additively combining steering vectors on a set of tasks. These results highlight conceptors as a promising tool for more effective steering of LLMs. Our code is available on github.com/jorispos/conceptorsteering.
comment: Presented at the MINT workshop at NeurIPS 2024
♻ ☆ Rethinking Decoders for Transformer-based Semantic Segmentation: A Compression Perspective NeurIPS2024
State-of-the-art methods for Transformer-based semantic segmentation typically adopt Transformer decoders that are used to extract additional embeddings from image embeddings via cross-attention, refine either or both types of embeddings via self-attention, and project image embeddings onto the additional embeddings via dot-product. Despite their remarkable success, these empirical designs still lack theoretical justifications or interpretations, thus hindering potentially principled improvements. In this paper, we argue that there are fundamental connections between semantic segmentation and compression, especially between the Transformer decoders and Principal Component Analysis (PCA). From such a perspective, we derive a white-box, fully attentional DEcoder for PrIncipled semantiC segemenTation (DEPICT), with the interpretations as follows: 1) the self-attention operator refines image embeddings to construct an ideal principal subspace that aligns with the supervision and retains most information; 2) the cross-attention operator seeks to find a low-rank approximation of the refined image embeddings, which is expected to be a set of orthonormal bases of the principal subspace and corresponds to the predefined classes; 3) the dot-product operation yields compact representation for image embeddings as segmentation masks. Experiments conducted on dataset ADE20K find that DEPICT consistently outperforms its black-box counterpart, Segmenter, and it is light weight and more robust.
comment: NeurIPS2024. Code:https://github.com/QishuaiWen/DEPICT/
♻ ☆ Automation of Quantum Dot Measurement Analysis via Explainable Machine Learning AAAI 2024
The rapid development of quantum dot (QD) devices for quantum computing has necessitated more efficient and automated methods for device characterization and tuning. This work demonstrates the feasibility and advantages of applying explainable machine learning techniques to the analysis of quantum dot measurements, paving the way for further advances in automated and transparent QD device tuning. Many of the measurements acquired during the tuning process come in the form of images that need to be properly analyzed to guide the subsequent tuning steps. By design, features present in such images capture certain behaviors or states of the measured QD devices. When considered carefully, such features can aid the control and calibration of QD devices. An important example of such images are so-called $\textit{triangle plots}$, which visually represent current flow and reveal characteristics important for QD device calibration. While image-based classification tools, such as convolutional neural networks (CNNs), can be used to verify whether a given measurement is $\textit{good}$ and thus warrants the initiation of the next phase of tuning, they do not provide any insights into how the device should be adjusted in the case of $\textit{bad}$ images. This is because CNNs sacrifice prediction and model intelligibility for high accuracy. To ameliorate this trade-off, a recent study introduced an image vectorization approach that relies on the Gabor wavelet transform (Schug $\textit{et al.}$ 2024 $\textit{Proc. XAI4Sci: Explainable Machine Learning for Sciences Workshop (AAAI 2024) (Vancouver, Canada)}$ pp 1-6). Here we propose an alternative vectorization method that involves mathematical modeling of synthetic triangles to mimic the experimental data. Using explainable boosting machines, we show that this new method offers superior explainability of model prediction without sacrificing accuracy.
comment: 20 pages, 5 figures, abbreviated version published in Proceedings of the XAI4Sci: Explainable machine learning for sciences workshop at AAAI 2024, (Vancouver, Canada)
♻ ☆ Explainable AI for Classifying UTI Risk Groups Using a Real-World Linked EHR and Pathology Lab Dataset
The use of machine learning and AI on electronic health records (EHRs) holds substantial potential for clinical insight. However, this approach faces challenges due to data heterogeneity, sparsity, temporal misalignment, and limited labeled outcomes. In this context, we leverage a linked EHR dataset of approximately one million de-identified individuals from Bristol, North Somerset, and South Gloucestershire, UK, to characterize urinary tract infections (UTIs). We implemented a data pre-processing and curation pipeline that transforms the raw EHR data into a structured format suitable for developing predictive models focused on data fairness, accountability and transparency. Given the limited availability and biases of ground truth UTI outcomes, we introduce a UTI risk estimation framework informed by clinical expertise to estimate UTI risk across individual patient timelines. Pairwise XGBoost models are trained using this framework to differentiate UTI risk categories with explainable AI techniques applied to identify key predictors and support interpretability. Our findings reveal differences in clinical and demographic predictors across risk groups. While this study highlights the potential of AI-driven insights to support UTI clinical decision-making, further investigation of patient sub-strata and extensive validation are needed to ensure robustness and applicability in clinical practice.
♻ ☆ Light Transport-aware Diffusion Posterior Sampling for Single-View Reconstruction of 3D Volumes
We introduce a single-view reconstruction technique of volumetric fields in which multiple light scattering effects are omnipresent, such as in clouds. We model the unknown distribution of volumetric fields using an unconditional diffusion model trained on a novel benchmark dataset comprising 1,000 synthetically simulated volumetric density fields. The neural diffusion model is trained on the latent codes of a novel, diffusion-friendly, monoplanar representation. The generative model is used to incorporate a tailored parametric diffusion posterior sampling technique into different reconstruction tasks. A physically-based differentiable volume renderer is employed to provide gradients with respect to light transport in the latent space. This stands in contrast to classic NeRF approaches and makes the reconstructions better aligned with observed data. Through various experiments, we demonstrate single-view reconstruction of volumetric clouds at a previously unattainable quality.
♻ ☆ Towards an Information Theoretic Framework of Context-Based Offline Meta-Reinforcement Learning
As a marriage between offline RL and meta-RL, the advent of offline meta-reinforcement learning (OMRL) has shown great promise in enabling RL agents to multi-task and quickly adapt while acquiring knowledge safely. Among which, context-based OMRL (COMRL) as a popular paradigm, aims to learn a universal policy conditioned on effective task representations. In this work, by examining several key milestones in the field of COMRL, we propose to integrate these seemingly independent methodologies into a unified framework. Most importantly, we show that the pre-existing COMRL algorithms are essentially optimizing the same mutual information objective between the task variable $M$ and its latent representation $Z$ by implementing various approximate bounds. Such theoretical insight offers ample design freedom for novel algorithms. As demonstrations, we propose a supervised and a self-supervised implementation of $I(Z; M)$, and empirically show that the corresponding optimization algorithms exhibit remarkable generalization across a broad spectrum of RL benchmarks, context shift scenarios, data qualities and deep learning architectures. This work lays the information theoretic foundation for COMRL methods, leading to a better understanding of task representation learning in the context of reinforcement learning. Given its generality, we envision our framework as a promising offline pre-training paradigm of foundation models for decision making.
comment: 26 pages, 8 figures, 7 tables. TLDR: We propose a novel information theoretic framework of the context-based offline meta-RL paradigm, which unifies several mainstream methods and leads to two robust algorithm implementations
♻ ☆ Project Tracyn: Generative Artificial Intelligence based Peripherals Trace Synthesizer
Peripheral Component Interconnect Express (PCIe) is the de facto interconnect standard for high-speed peripherals and CPUs. Prototyping and optimizing PCIe devices for emerging scenarios is an ongoing challenge. Since Transaction Layer Packets (TLPs) capture device-CPU interactions, it is crucial to analyze and generate realistic TLP traces for effective device design and optimization. Generative AI offers a promising approach for creating intricate, custom TLP traces necessary for PCIe hardware and software development. However, existing models often generate impractical traces due to the absence of PCIe-specific constraints, such as TLP ordering and causality. This paper presents Phantom, the first framework that treats TLP trace generation as a generative AI problem while incorporating PCIe-specific constraints. We validate Phantom's effectiveness by generating TLP traces for an actual PCIe network interface card. Experimental results show that Phantom produces practical, large-scale TLP traces, significantly outperforming existing models, with improvements of up to 1000$\times$ in task-specific metrics and up to 2.19$\times$ in Frechet Inception Distance (FID) compared to backbone-only methods.
♻ ☆ Design of 2D Skyrmionic Metamaterial Through Controlled Assembly
Despite extensive research on magnetic skyrmions and antiskyrmions, a significant challenge remains in crafting nontrivial high-order skyrmionic textures with varying, or even tailor-made, topologies. We address this challenge, by focusing on a construction pathway of skyrmionic metamaterials within a monolayer thin film and suggest several skyrmionic metamaterials that are surprisingly stable, i.e., long-lived, due to a self-stabilization mechanism. This makes these new textures promising for applications. Central to our approach is the concept of 'simulated controlled assembly', in short, a protocol inspired by 'click chemistry' that allows for positioning topological magnetic structures where one likes, and then allowing for energy minimization to elucidate the stability. Utilizing high-throughput atomistic-spin-dynamic simulations alongside state-of-the-art AI-driven tools, we have isolated skyrmions (topological charge Q=1), antiskyrmions (Q=-1), and skyrmionium (Q=0). These entities serve as foundational 'skyrmionic building blocks' to form the here reported intricate textures. In this work, two key contributions are introduced to the field of skyrmionic systems. First, we present a a novel combination of atomistic spin dynamics simulations and controlled assembly protocols for the stabilization and investigation of new topological magnets. Second, using the aforementioned methods we report on the discovery of skyrmionic metamaterials.
♻ ☆ BayesAdapter: enhanced uncertainty estimation in CLIP few-shot adaptation
The emergence of large pre-trained vision-language models (VLMs) represents a paradigm shift in machine learning, with unprecedented results in a broad span of visual recognition tasks. CLIP, one of the most popular VLMs, has exhibited remarkable zero-shot and transfer learning capabilities in classification. To transfer CLIP to downstream tasks, adapters constitute a parameter-efficient approach that avoids backpropagation through the large model (unlike related prompt learning methods). However, CLIP adapters have been developed to target discriminative performance, and the quality of their uncertainty estimates has been overlooked. In this work we show that the discriminative performance of state-of-the-art CLIP adapters does not always correlate with their uncertainty estimation capabilities, which are essential for a safe deployment in real-world scenarios. We also demonstrate that one of such adapters is obtained through MAP inference from a more general probabilistic framework. Based on this observation we introduce BayesAdapter, which leverages Bayesian inference to estimate a full probability distribution instead of a single point, better capturing the variability inherent in the parameter space. In a comprehensive empirical evaluation we show that our approach obtains high quality uncertainty estimates in the predictions, standing out in calibration and selective classification. Our code will be publicly available upon acceptance of the paper.
comment: 30 pages, 5 figures, 23 tables
♻ ☆ Exploring energy minimization to model strain localization as a strong discontinuity using Physics Informed Neural Networks
We explore the possibilities of using energy minimization for the numerical modeling of strain localization in solids as a sharp discontinuity in the displacement field. For this purpose, we consider (regularized) strong discontinuity kinematics in elastoplastic solids. The corresponding mathematical model is discretized using Artificial Neural Networks (ANNs), aiming to predict both the magnitude and location of the displacement jump from energy minimization, $\textit{i.e.}$, within a variational setting. The architecture takes care of the kinematics, while the loss function takes care of the variational statement of the boundary value problem. The main idea behind this approach is to solve both the equilibrium problem and the location of the localization band by means of trainable parameters in the ANN. As a proof of concept, we show through both 1D and 2D numerical examples that the computational modeling of strain localization for elastoplastic solids using energy minimization is feasible.
♻ ☆ QuadWBG: Generalizable Quadrupedal Whole-Body Grasping
Legged robots with advanced manipulation capabilities have the potential to significantly improve household duties and urban maintenance. Despite considerable progress in developing robust locomotion and precise manipulation methods, seamlessly integrating these into cohesive whole-body control for real-world applications remains challenging. In this paper, we present a modular framework for robust and generalizable whole-body loco-manipulation controller based on a single arm-mounted camera. By using reinforcement learning (RL), we enable a robust low-level policy for command execution over 5 dimensions (5D) and a grasp-aware high-level policy guided by a novel metric, Generalized Oriented Reachability Map (GORM). The proposed system achieves state-of-the-art one-time grasping accuracy of 89% in the real world, including challenging tasks such as grasping transparent objects. Through extensive simulations and real-world experiments, we demonstrate that our system can effectively manage a large workspace, from floor level to above body height, and perform diverse whole-body loco-manipulation tasks.
♻ ☆ AI-Driven Early Mental Health Screening: Analyzing Selfies of Pregnant Women ALT
Major Depressive Disorder and anxiety disorders affect millions globally, contributing significantly to the burden of mental health issues. Early screening is crucial for effective intervention, as timely identification of mental health issues can significantly improve treatment outcomes. Artificial intelligence (AI) can be valuable for improving the screening of mental disorders, enabling early intervention and better treatment outcomes. AI-driven screening can leverage the analysis of multiple data sources, including facial features in digital images. However, existing methods often rely on controlled environments or specialized equipment, limiting their broad applicability. This study explores the potential of AI models for ubiquitous depression-anxiety screening given face-centric selfies. The investigation focuses on high-risk pregnant patients, a population that is particularly vulnerable to mental health issues. To cope with limited training data resulting from our clinical setup, pre-trained models were utilized in two different approaches: fine-tuning convolutional neural networks (CNNs) originally designed for facial expression recognition and employing vision-language models (VLMs) for zero-shot analysis of facial expressions. Experimental results indicate that the proposed VLM-based method significantly outperforms CNNs, achieving an accuracy of 77.6%. Although there is significant room for improvement, the results suggest that VLMs can be a promising approach for mental health screening.
comment: This article has been accepted for publication in HEALTHINF25 at the 18th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2025)
♻ ☆ Spectral complexity of deep neural networks
It is well-known that randomly initialized, push-forward, fully-connected neural networks weakly converge to isotropic Gaussian processes, in the limit where the width of all layers goes to infinity. In this paper, we propose to use the angular power spectrum of the limiting field to characterize the complexity of the network architecture. In particular, we define sequences of random variables associated with the angular power spectrum, and provide a full characterization of the network complexity in terms of the asymptotic distribution of these sequences as the depth diverges. On this basis, we classify neural networks as low-disorder, sparse, or high-disorder; we show how this classification highlights a number of distinct features for standard activation functions, and in particular, sparsity properties of ReLU networks. Our theoretical results are also validated by numerical simulations.
♻ ☆ Improving Forward Compatibility in Class Incremental Learning by Increasing Representation Rank and Feature Richness
Class Incremental Learning (CIL) constitutes a pivotal subfield within continual learning, aimed at enabling models to progressively learn new classification tasks while retaining knowledge obtained from prior tasks. Although previous studies have predominantly focused on backward compatible approaches to mitigate catastrophic forgetting, recent investigations have introduced forward compatible methods to enhance performance on novel tasks and complement existing backward compatible methods. In this study, we introduce an effective-Rank based Feature Richness enhancement (RFR) method, designed for improving forward compatibility. Specifically, this method increases the effective rank of representations during the base session, thereby facilitating the incorporation of more informative features pertinent to unseen novel tasks. Consequently, RFR achieves dual objectives in backward and forward compatibility: minimizing feature extractor modifications and enhancing novel task performance, respectively. To validate the efficacy of our approach, we establish a theoretical connection between effective rank and the Shannon entropy of representations. Subsequently, we conduct comprehensive experiments by integrating RFR into eleven well-known CIL methods. Our results demonstrate the effectiveness of our approach in enhancing novel-task performance while mitigating catastrophic forgetting. Furthermore, our method notably improves the average incremental accuracy across all eleven cases examined.
♻ ☆ QUACK: Quantum Aligned Centroid Kernel
Quantum computing (QC) seems to show potential for application in machine learning (ML). In particular quantum kernel methods (QKM) exhibit promising properties for use in supervised ML tasks. However, a major disadvantage of kernel methods is their unfavorable quadratic scaling with the number of training samples. Together with the limits imposed by currently available quantum hardware (NISQ devices) with their low qubit coherence times, small number of qubits, and high error rates, the use of QC in ML at an industrially relevant scale is currently impossible. As a small step in improving the potential applications of QKMs, we introduce QUACK, a quantum kernel algorithm whose time complexity scales linear with the number of samples during training, and independent of the number of training samples in the inference stage. In the training process, only the kernel entries for the samples and the centers of the classes are calculated, i.e. the maximum shape of the kernel for n samples and c classes is (n, c). During training, the parameters of the quantum kernel and the positions of the centroids are optimized iteratively. In the inference stage, for every new sample the circuit is only evaluated for every centroid, i.e. c times. We show that the QUACK algorithm nevertheless provides satisfactory results and can perform at a similar level as classical kernel methods with quadratic scaling during training. In addition, our (simulated) algorithm is able to handle high-dimensional datasets such as MNIST with 784 features without any dimensionality reduction.
comment: 2nd place Best Paper award in QML track @ IEEE International Conference on Quantum Computing and Engineering (QCE) 2024
♻ ☆ Quantifying Aleatoric Uncertainty of the Treatment Effect: A Novel Orthogonal Learner
Estimating causal quantities from observational data is crucial for understanding the safety and effectiveness of medical treatments. However, to make reliable inferences, medical practitioners require not only estimating averaged causal quantities, such as the conditional average treatment effect, but also understanding the randomness of the treatment effect as a random variable. This randomness is referred to as aleatoric uncertainty and is necessary for understanding the probability of benefit from treatment or quantiles of the treatment effect. Yet, the aleatoric uncertainty of the treatment effect has received surprisingly little attention in the causal machine learning community. To fill this gap, we aim to quantify the aleatoric uncertainty of the treatment effect at the covariate-conditional level, namely, the conditional distribution of the treatment effect (CDTE). Unlike average causal quantities, the CDTE is not point identifiable without strong additional assumptions. As a remedy, we employ partial identification to obtain sharp bounds on the CDTE and thereby quantify the aleatoric uncertainty of the treatment effect. We then develop a novel, orthogonal learner for the bounds on the CDTE, which we call AU-learner. We further show that our AU-learner has several strengths in that it satisfies Neyman-orthogonality and, thus, quasi-oracle efficiency. Finally, we propose a fully-parametric deep learning instantiation of our AU-learner.
♻ ☆ Benchmarking Counterfactual Image Generation NeurIPS 2024
Generative AI has revolutionised visual content editing, empowering users to effortlessly modify images and videos. However, not all edits are equal. To perform realistic edits in domains such as natural image or medical imaging, modifications must respect causal relationships inherent to the data generation process. Such image editing falls into the counterfactual image generation regime. Evaluating counterfactual image generation is substantially complex: not only it lacks observable ground truths, but also requires adherence to causal constraints. Although several counterfactual image generation methods and evaluation metrics exist, a comprehensive comparison within a unified setting is lacking. We present a comparison framework to thoroughly benchmark counterfactual image generation methods. We integrate all models that have been used for the task at hand and expand them to novel datasets and causal graphs, demonstrating the superiority of Hierarchical VAEs across most datasets and metrics. Our framework is implemented in a user-friendly Python package that can be extended to incorporate additional SCMs, causal methods, generative models, and datasets for the community to build on. Code: https://github.com/gulnazaki/counterfactual-benchmark.
comment: Published as a conference paper at NeurIPS 2024 Datasets and Benchmarks Track https://openreview.net/forum?id=0T8xRFrScB Project page: https://gulnazaki.github.io/counterfactual-benchmark
♻ ☆ Imitating from auxiliary imperfect demonstrations via Adversarial Density Weighted Regression
We propose a novel one-step supervised imitation learning (IL) framework called Adversarial Density Regression (ADR). This IL framework aims to correct the policy learned on unknown-quality to match the expert distribution by utilizing demonstrations, without relying on the Bellman operator. Specifically, ADR addresses several limitations in previous IL algorithms: First, most IL algorithms are based on the Bellman operator, which inevitably suffer from cumulative offsets from sub-optimal rewards during multi-step update processes. Additionally, off-policy training frameworks suffer from Out-of-Distribution (OOD) state-actions. Second, while conservative terms help solve the OOD issue, balancing the conservative term is difficult. To address these limitations, we fully integrate a one-step density-weighted Behavioral Cloning (BC) objective for IL with auxiliary imperfect demonstration. Theoretically, we demonstrate that this adaptation can effectively correct the distribution of policies trained on unknown-quality datasets to align with the expert policy's distribution. Moreover, the difference between the empirical and the optimal value function is proportional to the upper bound of ADR's objective, indicating that minimizing ADR's objective is akin to approaching the optimal value. Experimentally, we validated the performance of ADR by conducting extensive evaluations. Specifically, ADR outperforms all of the selected IL algorithms on tasks from the Gym-Mujoco domain. Meanwhile, it achieves an 89.5% improvement over IQL when utilizing ground truth rewards on tasks from the Adroit and Kitchen domains. Our codebase will be released at: https://github.com/stevezhangzA/Adverserial_Density_Regression.
♻ ☆ D3RM: A Discrete Denoising Diffusion Refinement Model for Piano Transcription ICASSP 2025
Diffusion models have been widely used in the generative domain due to their convincing performance in modeling complex data distributions. Moreover, they have shown competitive results on discriminative tasks, such as image segmentation. While diffusion models have also been explored for automatic music transcription, their performance has yet to reach a competitive level. In this paper, we focus on discrete diffusion model's refinement capabilities and present a novel architecture for piano transcription. Our model utilizes Neighborhood Attention layers as the denoising module, gradually predicting the target high-resolution piano roll, conditioned on the finetuned features of a pretrained acoustic model. To further enhance refinement, we devise a novel strategy which applies distinct transition states during training and inference stage of discrete diffusion models. Experiments on the MAESTRO dataset show that our approach outperforms previous diffusion-based piano transcription models and the baseline model in terms of F1 score. Our code is available in https://github.com/hanshounsu/d3rm.
comment: Accepted to ICASSP 2025
♻ ☆ Are LLMs Good Cryptic Crossword Solvers?
Cryptic crosswords are puzzles that rely not only on general knowledge but also on the solver's ability to manipulate language on different levels and deal with various types of wordplay. Previous research suggests that solving such puzzles is a challenge even for modern NLP models. However, the abilities of large language models (LLMs) have not yet been tested on this task. In this paper, we establish the benchmark results for three popular LLMs -- LLaMA2, Mistral, and ChatGPT -- showing that their performance on this task is still far from that of humans.
♻ ☆ SyncDiff: Synchronized Motion Diffusion for Multi-Body Human-Object Interaction Synthesis
Synthesizing realistic human-object interaction motions is a critical problem in VR/AR and human animation. Unlike the commonly studied scenarios involving a single human or hand interacting with one object, we address a more generic multi-body setting with arbitrary numbers of humans, hands, and objects. This complexity introduces significant challenges in synchronizing motions due to the high correlations and mutual influences among bodies. To address these challenges, we introduce SyncDiff, a novel method for multi-body interaction synthesis using a synchronized motion diffusion strategy. SyncDiff employs a single diffusion model to capture the joint distribution of multi-body motions. To enhance motion fidelity, we propose a frequency-domain motion decomposition scheme. Additionally, we introduce a new set of alignment scores to emphasize the synchronization of different body motions. SyncDiff jointly optimizes both data sample likelihood and alignment likelihood through an explicit synchronization strategy. Extensive experiments across four datasets with various multi-body configurations demonstrate the superiority of SyncDiff over existing state-of-the-art motion synthesis methods.
♻ ☆ Initialization is Critical to Whether Transformers Fit Composite Functions by Reasoning or Memorizing
Transformers have shown impressive capabilities across various tasks, but their performance on compositional problems remains a topic of debate. In this work, we investigate the mechanisms of how transformers behave on unseen compositional tasks. We discover that the parameter initialization scale plays a critical role in determining whether the model learns inferential (reasoning-based) solutions, which capture the underlying compositional primitives, or symmetric (memory-based) solutions, which simply memorize mappings without understanding the compositional structure. By analyzing the information flow and vector representations within the model, we reveal the distinct mechanisms underlying these solution types. We further find that inferential (reasoning-based) solutions exhibit low complexity bias, which we hypothesize is a key factor enabling them to learn individual mappings for single anchors. We validate our conclusions on various real-world datasets. Our findings provide valuable insights into the role of initialization scale in tuning the reasoning and memorizing ability and we propose the initialization rate $\gamma$ to be a convenient tunable hyper-parameter in common deep learning frameworks, where $1/d_{\mathrm{in}}^\gamma$ is the standard deviation of parameters of the layer with $d_{\mathrm{in}}$ input neurons.
♻ ☆ GFairHint: Improving Individual Fairness for Graph Neural Networks via Fairness Hint KDD 2025
Given the growing concerns about fairness in machine learning and the impressive performance of Graph Neural Networks (GNNs) on graph data learning, algorithmic fairness in GNNs has attracted significant attention. While many existing studies improve fairness at the group level, only a few works promote individual fairness, which renders similar outcomes for similar individuals. A desirable framework that promotes individual fairness should (1) balance between fairness and performance, (2) accommodate two commonly-used individual similarity measures (externally annotated and computed from input features), (3) generalize across various GNN models, and (4) be computationally efficient. Unfortunately, none of the prior work achieves all the desirables. In this work, we propose a novel method, GFairHint, which promotes individual fairness in GNNs and achieves all aforementioned desirables. GFairHint learns fairness representations through an auxiliary link prediction task, and then concatenates the representations with the learned node embeddings in original GNNs as a "fairness hint". Through extensive experimental investigations on five real-world graph datasets under three prevalent GNN models covering both individual similarity measures above, GFairHint achieves the best fairness results in almost all combinations of datasets with various backbone models, while generating comparable utility results, with much less computational cost compared to the previous state-of-the-art (SoTA) method.
comment: Accepted by the ACM Transactions on Knowledge Discovery from Data (TKDD 2025)
♻ ☆ CoNOAir: A Neural Operator for Forecasting Carbon Monoxide Evolution in Cities
Carbon Monoxide (CO) is a dominant pollutant in urban areas due to the energy generation from fossil fuels for industry, automobile, and domestic requirements. Forecasting the evolution of CO in real-time can enable the deployment of effective early warning systems and intervention strategies. However, the computational cost associated with the physics and chemistry-based simulation makes it prohibitive to implement such a model at the city and country scale. To address this challenge, here, we present a machine learning model based on neural operator, namely, Complex Neural Operator for Air Quality (CoNOAir), that can effectively forecast CO concentrations. We demonstrate this by developing a country-level model for short-term (hourly) and long-term (72-hour) forecasts of CO concentrations. Our model outperforms state-of-the-art models such as Fourier neural operators (FNO) and provides reliable predictions for both short and long-term forecasts. We further analyse the capability of the model to capture extreme events and generate forecasts in urban cities in India. Interestingly, we observe that the model predicts the next hour CO concentrations with R2 values greater than 0.95 for all the cities considered. The deployment of such a model can greatly assist the governing bodies to provide early warning, plan intervention strategies, and develop effective strategies by considering several what-if scenarios. Altogether, the present approach could provide a fillip to real-time predictions of CO pollution in urban cities.
comment: 28 pages, 14 figures, under submission process
♻ ☆ A monthly sub-national Harmonized Food Insecurity Dataset for comprehensive analysis and predictive modeling
Food security is a complex, multidimensional concept challenging to measure comprehensively. Effective anticipation, monitoring, and mitigation of food crises require timely and comprehensive global data. This paper introduces the Harmonized Food Insecurity Dataset (HFID), an open-source resource consolidating four key data sources: the Integrated Food Security Phase Classification (IPC)/Cadre Harmonis\'e (CH) phases, the Famine Early Warning Systems Network (FEWS NET) IPC-compatible phases, and the World Food Program's (WFP) Food Consumption Score (FCS) and reduced Coping Strategy Index (rCSI). Updated monthly and using a common reference system for administrative units, the HFID offers extensive spatial and temporal coverage. It serves as a vital tool for food security experts and humanitarian agencies, providing a unified resource for analyzing food security conditions and highlighting global data disparities. The scientific community can also leverage the HFID to develop data-driven predictive models, enhancing the capacity to forecast and prevent future food crises.
comment: The authors Melissande Machefer and Michele Ronco have contributed equally as both first authors to this work. This work is currently being reviewed in a peer-reviewed journal
♻ ☆ MusicLIME: Explainable Multimodal Music Understanding ICASSP 2025
Multimodal models are critical for music understanding tasks, as they capture the complex interplay between audio and lyrics. However, as these models become more prevalent, the need for explainability grows-understanding how these systems make decisions is vital for ensuring fairness, reducing bias, and fostering trust. In this paper, we introduce MusicLIME, a model-agnostic feature importance explanation method designed for multimodal music models. Unlike traditional unimodal methods, which analyze each modality separately without considering the interaction between them, often leading to incomplete or misleading explanations, MusicLIME reveals how audio and lyrical features interact and contribute to predictions, providing a holistic view of the model's decision-making. Additionally, we enhance local explanations by aggregating them into global explanations, giving users a broader perspective of model behavior. Through this work, we contribute to improving the interpretability of multimodal music models, empowering users to make informed choices, and fostering more equitable, fair, and transparent music understanding systems.
comment: GitHub repository: https://github.com/IamTheo2000/MusicLIME. To be presented at ICASSP 2025
♻ ☆ CAB: Comprehensive Attention Benchmarking on Long Sequence Modeling
Transformer has achieved remarkable success in language, image, and speech processing. Recently, various efficient attention architectures have been proposed to improve transformer's efficiency while largely preserving its efficacy, especially in modeling long sequences. A widely-used benchmark to test these efficient methods' capability on long-range modeling is Long Range Arena (LRA). However, LRA only focuses on the standard bidirectional (or noncausal) self attention, and completely ignores cross attentions and unidirectional (or causal) attentions, which are equally important to downstream applications. In this paper, we propose Comprehensive Attention Benchmark (CAB) under a fine-grained attention taxonomy with four distinguishable attention patterns, namely, noncausal self, causal self, noncausal cross, and causal cross attentions. CAB collects seven real-world tasks from different research areas to evaluate efficient attentions under the four attention patterns. Among these tasks, CAB validates efficient attentions in eight backbone networks to show their generalization across neural architectures. We conduct exhaustive experiments to benchmark the performances of nine widely-used efficient attention architectures designed with different philosophies on CAB. Extensive experimental results also shed light on the fundamental problems of efficient attentions, such as efficiency length against vanilla attention, performance consistency across attention patterns, the benefit of attention mechanisms, and interpolation/extrapolation on long-context language modeling.
Amortizing intractable inference in diffusion models for vision, language, and control NeurIPS 2024
Diffusion models have emerged as effective distribution estimators in vision, language, and reinforcement learning, but their use as priors in downstream tasks poses an intractable posterior inference problem. This paper studies amortized sampling of the posterior over data, $\mathbf{x}\sim p^{\rm post}(\mathbf{x})\propto p(\mathbf{x})r(\mathbf{x})$, in a model that consists of a diffusion generative model prior $p(\mathbf{x})$ and a black-box constraint or likelihood function $r(\mathbf{x})$. We state and prove the asymptotic correctness of a data-free learning objective, relative trajectory balance, for training a diffusion model that samples from this posterior, a problem that existing methods solve only approximately or in restricted cases. Relative trajectory balance arises from the generative flow network perspective on diffusion models, which allows the use of deep reinforcement learning techniques to improve mode coverage. Experiments illustrate the broad potential of unbiased inference of arbitrary posteriors under diffusion priors: in vision (classifier guidance), language (infilling under a discrete diffusion LLM), and multimodal data (text-to-image generation). Beyond generative modeling, we apply relative trajectory balance to the problem of continuous control with a score-based behavior prior, achieving state-of-the-art results on benchmarks in offline reinforcement learning.
comment: NeurIPS 2024; code: https://github.com/GFNOrg/diffusion-finetuning
♻ ☆ Efficient Large Foundation Models Design: A Perspective From Model and System Co-Design
This paper focuses on modern efficient training and inference technologies on foundation models and illustrates them from two perspectives: model and system design. Model and System Design optimize LLM training and inference from different aspects to save computational resources, making LLMs more efficient, affordable, and more accessible. The paper list repository is available at \url{https://github.com/NoakLiu/Efficient-Foundation-Models-Survey}
♻ ☆ EVA-S2PLoR: A Secure Element-wise Multiplication Meets Logistic Regression on Heterogeneous Database
Accurate nonlinear computation is a key challenge in privacy-preserving machine learning (PPML). Most existing frameworks approximate it through linear operations, resulting in significant precision loss. This paper proposes an efficient, verifiable and accurate security 2-party logistic regression framework (EVA-S2PLoR), which achieves accurate nonlinear function computation through a novel secure element-wise multiplication protocol and its derived protocols. Our framework primarily includes secure 2-party vector element-wise multiplication, addition to multiplication, reciprocal, and sigmoid function based on data disguising technology, where high efficiency and accuracy are guaranteed by the simple computation flow based on the real number domain and the few number of fixed communication rounds. We provide secure and robust anomaly detection through dimension transformation and Monte Carlo methods. EVA-S2PLoR outperforms many advanced frameworks in terms of precision (improving the performance of the sigmoid function by about 10 orders of magnitude compared to most frameworks) and delivers the best overall performance in secure logistic regression experiments.
♻ ☆ Optimally Solving Simultaneous-Move Dec-POMDPs: The Sequential Central Planning Approach
The centralized training for decentralized execution paradigm emerged as the state-of-the-art approach to $\epsilon$-optimally solving decentralized partially observable Markov decision processes. However, scalability remains a significant issue. This paper presents a novel and more scalable alternative, namely the sequential-move centralized training for decentralized execution. This paradigm further pushes the applicability of the Bellman's principle of optimality, raising three new properties. First, it allows a central planner to reason upon sufficient sequential-move statistics instead of prior simultaneous-move ones. Next, it proves that $\epsilon$-optimal value functions are piecewise linear and convex in such sufficient sequential-move statistics. Finally, it drops the complexity of the backup operators from double exponential to polynomial at the expense of longer planning horizons. Besides, it makes it easy to use single-agent methods, e.g., SARSA algorithm enhanced with these findings, while still preserving convergence guarantees. Experiments on two- as well as many-agent domains from the literature against $\epsilon$-optimal simultaneous-move solvers confirm the superiority of our novel approach. This paradigm opens the door for efficient planning and reinforcement learning methods for multi-agent systems.
♻ ☆ Exploring Feature-based Knowledge Distillation for Recommender System: A Frequency Perspective KDD 2025
In this paper, we analyze the feature-based knowledge distillation for recommendation from the frequency perspective. By defining knowledge as different frequency components of the features, we theoretically demonstrate that regular feature-based knowledge distillation is equivalent to equally minimizing losses on all knowledge and further analyze how this equal loss weight allocation method leads to important knowledge being overlooked. In light of this, we propose to emphasize important knowledge by redistributing knowledge weights. Furthermore, we propose FreqD, a lightweight knowledge reweighting method, to avoid the computational cost of calculating losses on each knowledge. Extensive experiments demonstrate that FreqD consistently and significantly outperforms state-of-the-art knowledge distillation methods for recommender systems. Our code is available at https://github.com/woriazzc/KDs.
comment: ACM KDD 2025 Accepted
♻ ☆ Explainable Metrics for the Assessment of Neurodegenerative Diseases through Handwriting Analysis
Motor dysfunction is a common sign of neurodegenerative diseases (NDs) such as Parkinson's disease (PD) and Alzheimer's disease (AD), but may be difficult to detect, especially in the early stages. In this work, we examine the behavior of a wide array of explainable metrics extracted from the handwriting signals of 113 subjects performing multiple tasks on a digital tablet, as part of the Neurological Signals dataset. The aim is to measure their effectiveness in characterizing NDs, including AD and PD. To this end, task-agnostic and task-specific metrics are extracted from 14 distinct tasks. Subsequently, through statistical analysis and a series of classification experiments, we investigate which metrics provide greater discriminative power between NDs and healthy controls and amongst different NDs. Preliminary results indicate that the tasks at hand can all be effectively leveraged to distinguish between the considered set of NDs, specifically by measuring the stability, the speed of writing, the time spent not writing, and the pressure variations between groups from our handcrafted explainable metrics, which shows p-values lower than 0.0001 for multiple tasks. Using various binary classification algorithms on the computed metrics, we obtain up to 87 % accuracy for the discrimination between AD and healthy controls (CTL), and up to 69 % for the discrimination between PD and CTL.
comment: 14 pages including references, under review in IEEE JHBI
♻ ☆ An empirical study of LLaMA3 quantization: from LLMs to MLLMs
The LLaMA family, a collection of foundation language models ranging from 7B to 65B parameters, has become one of the most powerful open-source large language models (LLMs) and the popular LLM backbone of multi-modal large language models (MLLMs), widely used in computer vision and natural language understanding tasks. In particular, LLaMA3 models have recently been released and have achieved impressive performance in various domains with super-large scale pre-training on over 15T tokens of data. Given the wide application of low-bit quantization for LLMs in resource-constrained scenarios, we explore LLaMA3's capabilities when quantized to low bit-width. This exploration can potentially provide new insights and challenges for the low-bit quantization of LLaMA3 and other future LLMs, especially in addressing performance degradation issues that suffer in LLM compression. Specifically, we comprehensively evaluate the 10 existing post-training quantization and LoRA fine-tuning (LoRA-FT) methods of LLaMA3 on 1-8 bits and various datasets to reveal the low-bit quantization performance of LLaMA3. To uncover the capabilities of low-bit quantized MLLM, we assessed the performance of the LLaMA3-based LLaVA-Next-8B model under 2-4 ultra-low bits with post-training quantization methods. Our experimental results indicate that LLaMA3 still suffers from non-negligible degradation in linguistic and visual contexts, particularly under ultra-low bit widths. This highlights the significant performance gap at low bit-width that needs to be addressed in future developments. We expect that this empirical study will prove valuable in advancing future models, driving LLMs and MLLMs to achieve higher accuracy at lower bit to enhance practicality. Our project is released on https://github.com/Macaronlin/LLaMA3-Quantization , and quantized models are released at https://huggingface.co/Efficient-ML .
♻ ☆ Continual Learning with Strategic Selection and Forgetting for Network Intrusion Detection
Intrusion Detection Systems (IDS) are crucial for safeguarding digital infrastructure. In dynamic network environments, both threat landscapes and normal operational behaviors are constantly changing, resulting in concept drift. While continuous learning mitigates the adverse effects of concept drift, insufficient attention to drift patterns and excessive preservation of outdated knowledge can still hinder the IDS's adaptability. In this paper, we propose SSF (Strategic Selection and Forgetting), a novel continual learning method for IDS, providing continuous model updates with a constantly refreshed memory buffer. Our approach features a strategic sample selection algorithm to select representative new samples and a strategic forgetting mechanism to drop outdated samples. The proposed strategic sample selection algorithm prioritizes new samples that cause the `drifted' pattern, enabling the model to better understand the evolving landscape. Additionally, we introduce strategic forgetting upon detecting significant drift by discarding outdated samples to free up memory, allowing the incorporation of more recent data. SSF captures evolving patterns effectively and ensures the model is aligned with the change of data patterns, significantly enhancing the IDS's adaptability to concept drift. The state-of-the-art performance of SSF on NSL-KDD and UNSW-NB15 datasets demonstrates its superior adaptability to concept drift for network intrusion detection.
comment: Accepted by IEEE International Conference on Computer Communications (INFOCOM) 2025
♻ ☆ Model-Agnostic Cosmological Inference with SDSS-IV eBOSS: Simultaneous Probing for Background and Perturbed Universe
Here we explore certain subtle features imprinted in data from the completed Sloan Digital Sky Survey IV (SDSS-IV) extended Baryon Oscillation Spectroscopic Survey (eBOSS) as a combined probe for the background and perturbed Universe. We reconstruct the baryon Acoustic Oscillation (BAO) and Redshift Space Distortion (RSD) observables as functions of redshift, using measurements from SDSS alone. We apply the Multi-Task Gaussian Process (MTGP) framework to model the interdependencies of cosmological observables $D_M(z)/r_d$, $D_H(z)/r_d$, and $f\sigma_8(z)$, and track their evolution across different redshifts. Subsequently, we obtain constrained three-dimensional phase space containing $D_M(z)/r_d$, $D_H(z)/r_d$, and $f\sigma_8(z)$ at different redshifts probed by the SDSS-IV eBOSS survey. Furthermore, assuming the $\Lambda$CDM model, we obtain constraints on model parameters $\Omega_{m}$, $H_{0}r_{d}$, $\sigma_{8}$ and $S_{8}$ at each redshift probed by SDSS-IV eBOSS. This indicates redshift-dependent trends in $H_0$, $\Omega_m$, $\sigma_8$ and $S_8$ in the $\Lambda$CDM model, suggesting a possible inconsistency in the $\Lambda$CDM model. Ours is a template for model-independent extraction of information for both background and perturbed Universe using a single galaxy survey taking into account all the existing correlations between background and perturbed observables and this can be easily extended to future DESI-3YR as well as Euclid results.
comment: 14 pages, 7 sets of figures, 3 tables. Comments are welcome. New references added
♻ ☆ AdaPRL: Adaptive Pairwise Regression Learning with Uncertainty Estimation for Universal Regression Tasks
Current deep regression models usually learn in point-wise way that treat each sample as an independent input, neglecting the relative ordering among different data. Consequently, the regression model could neglect the data 's interrelationships, potentially resulting in suboptimal performance. Moreover, the existence of aleatoric uncertainty in the training data may drive the model to capture non-generalizable patterns, contributing to increased overfitting. To address these issues, we propose a novel adaptive pairwise learning framework (AdaPRL) for regression tasks which leverages the relative differences between data points and integrates with deep probabilistic models to quantify the uncertainty associated with the predictions. Additionally, we adapt AdaPRL for applications in multi-task learning and multivariate time series forecasting. Extensive experiments with several real-world regression datasets including recommendation systems, age estimation, time series forecasting, natural language understanding, finance, and industry datasets show that AdaPRL is compatible with different backbone networks in various tasks and achieves state-of-the-art performance on the vast majority of tasks, highlighting its notable potential including enhancing prediction accuracy and ranking ability, increasing generalization capability, improving robustness to noisy data, improving resilience to reduced data, and enhancing interpretability, etc.
comment: 22 pages, 11 figures
♻ ☆ MIO: A Foundation Model on Multimodal Tokens
In this paper, we introduce MIO, a novel foundation model built on multimodal tokens, capable of understanding and generating speech, text, images, and videos in an end-to-end, autoregressive manner. While the emergence of large language models (LLMs) and multimodal large language models (MM-LLMs) propels advancements in artificial general intelligence through their versatile capabilities, they still lack true any-to-any understanding and generation. Recently, the release of GPT-4o has showcased the remarkable potential of any-to-any LLMs for complex real-world tasks, enabling omnidirectional input and output across images, speech, and text. However, it is closed-source and does not support the generation of multimodal interleaved sequences. To address this gap, we present MIO, which is trained on a mixture of discrete tokens across four modalities using causal multimodal modeling. MIO undergoes a four-stage training process: (1) alignment pre-training, (2) interleaved pre-training, (3) speech-enhanced pre-training, and (4) comprehensive supervised fine-tuning on diverse textual, visual, and speech tasks. Our experimental results indicate that MIO exhibits competitive, and in some cases superior, performance compared to previous dual-modal baselines, any-to-any model baselines, and even modality-specific baselines. Moreover, MIO demonstrates advanced capabilities inherent to its any-to-any feature, such as interleaved video-text generation, chain-of-visual-thought reasoning, visual guideline generation, instructional image editing, etc.
comment: Technical Report. Codes and models are available in https://github.com/MIO-Team/MIO
♻ ☆ Simplifying CLIP: Unleashing the Power of Large-Scale Models on Consumer-level Computers
Contrastive Language-Image Pre-training (CLIP) has attracted a surge of attention for its superior zero-shot performance and excellent transferability to downstream tasks. However, training such large-scale models usually requires substantial computation and storage, which poses barriers for general users with consumer-level computers. Motivated by this observation, in this paper we investigate how to achieve competitive performance on only one Nvidia RTX3090 GPU and with one terabyte for storing dataset. On one hand, we simplify the transformer block structure and combine Weight Inheritance with multi-stage Knowledge Distillation (WIKD), thereby reducing the parameters and improving the inference speed during training along with deployment. On the other hand, confronted with the convergence challenge posed by small dataset, we generate synthetic captions for each sample as data augmentation, and devise a novel Pair Matching (PM) loss to fully exploit the distinguishment among positive and negative image-text pairs. Extensive experiments demonstrate that our model can achieve a new state-of-the-art datascale-parameter-accuracy tradeoff, which could further popularize the CLIP model in the related research community.
♻ ☆ Buster: Implanting Semantic Backdoor into Text Encoder to Mitigate NSFW Content Generation
The rise of deep learning models in the digital era has raised substantial concerns regarding the generation of Not-Safe-for-Work (NSFW) content. Existing defense methods primarily involve model fine-tuning and post-hoc content moderation. Nevertheless, these approaches largely lack scalability in eliminating harmful content, degrade the quality of benign image generation, or incur high inference costs. To address these challenges, we propose an innovative framework named \textit{Buster}, which injects backdoors into the text encoder to prevent NSFW content generation. Buster leverages deep semantic information rather than explicit prompts as triggers, redirecting NSFW prompts towards targeted benign prompts. Additionally, Buster employs energy-based training data generation through Langevin dynamics for adversarial knowledge augmentation, thereby ensuring robustness in harmful concept definition. This approach demonstrates exceptional resilience and scalability in mitigating NSFW content. Particularly, Buster fine-tunes the text encoder of Text-to-Image models within merely five minutes, showcasing its efficiency. Our extensive experiments denote that Buster outperforms nine state-of-the-art baselines, achieving a superior NSFW content removal rate of at least 91.2\% while preserving the quality of harmless images.
♻ ☆ Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability
Mathematical reasoning tasks pose significant challenges for large language models (LLMs) because they require precise logical deduction and sequence analysis. In this work, we introduce the concept of critical tokens -- elements within reasoning trajectories that significantly influence incorrect outcomes. We present a novel framework for identifying these tokens through rollout sampling and demonstrate their substantial divergence from traditional error tokens. Through extensive experiments on datasets such as GSM8K and MATH500, we show that identifying and replacing critical tokens significantly improves model accuracy. We propose an efficient methodology for pinpointing these tokens in large-scale datasets using contrastive estimation and extend this framework to enhance model training processes with direct preference optimization (DPO). Experimental results on GSM8K and MATH500 benchmarks with the widely used models Llama-3 (8B and 70B) and Deepseek-math (7B) demonstrate the effectiveness of the proposed approach, cDPO. Our results underscore the potential of leveraging critical tokens to reduce errors in reasoning tasks, advancing the development of AI systems capable of robust logical deduction. Our code, annotated datasets, and trained models are available at https://github.com/chenzhiling9954/Critical-Tokens-Matter to support and encourage future research in this promising field.
comment: Work in progress
♻ ☆ Fast and reliable uncertainty quantification with neural network ensembles for industrial image classification
Image classification with neural networks (NNs) is widely used in industrial processes, situations where the model likely encounters unknown objects during deployment, i.e., out-of-distribution (OOD) data. Worryingly, NNs tend to make confident yet incorrect predictions when confronted with OOD data. To increase the models' reliability, they should quantify the uncertainty in their own predictions, communicating when the output should (not) be trusted. Deep ensembles, composed of multiple independent NNs, have been shown to perform strongly but are computationally expensive. Recent research has proposed more efficient NN ensembles, namely the snapshot, batch, and multi-input multi-output ensemble. This study investigates the predictive and uncertainty performance of efficient NN ensembles in the context of image classification for industrial processes. It is the first to provide a comprehensive comparison and it proposes a novel Diversity Quality metric to quantify the ensembles' performance on the in-distribution and OOD sets in one single metric. The results highlight the batch ensemble as a cost-effective and competitive alternative to the deep ensemble. It matches the deep ensemble in both uncertainty and accuracy while exhibiting considerable savings in training time, test time, and memory storage.
comment: Accepted Manuscript version of an article published in Annals of Operations Research
♻ ☆ Generalizing Weather Forecast to Fine-grained Temporal Scales via Physics-AI Hybrid Modeling
Data-driven artificial intelligence (AI) models have made significant advancements in weather forecasting, particularly in medium-range and nowcasting. However, most data-driven weather forecasting models are black-box systems that focus on learning data mapping rather than fine-grained physical evolution in the time dimension. Consequently, the limitations in the temporal scale of datasets prevent these models from forecasting at finer time scales. This paper proposes a physics-AI hybrid model (i.e., WeatherGFT) which generalizes weather forecasts to finer-grained temporal scales beyond training dataset. Specifically, we employ a carefully designed PDE kernel to simulate physical evolution on a small time scale (e.g., 300 seconds) and use a parallel neural networks with a learnable router for bias correction. Furthermore, we introduce a lead time-aware training framework to promote the generalization of the model at different lead times. The weight analysis of physics-AI modules indicates that physics conducts major evolution while AI performs corrections adaptively. Extensive experiments show that WeatherGFT trained on an hourly dataset, effectively generalizes forecasts across multiple time scales, including 30-minute, which is even smaller than the dataset's temporal resolution.
♻ ☆ On the Convergence of Continual Federated Learning Using Incrementally Aggregated Gradients
The holy grail of machine learning is to enable Continual Federated Learning (CFL) to enhance the efficiency, privacy, and scalability of AI systems while learning from streaming data. The primary challenge of a CFL system is to overcome global catastrophic forgetting, wherein the accuracy of the global model trained on new tasks declines on the old tasks. In this work, we propose Continual Federated Learning with Aggregated Gradients (C-FLAG), a novel replay-memory based federated strategy consisting of edge-based gradient updates on memory and aggregated gradients on the current data. We provide convergence analysis of the C-FLAG approach which addresses forgetting and bias while converging at a rate of $O(1/\sqrt{T})$ over $T$ communication rounds. We formulate an optimization sub-problem that minimizes catastrophic forgetting, translating CFL into an iterative algorithm with adaptive learning rates that ensure seamless learning across tasks. We empirically show that C-FLAG outperforms several state-of-the-art baselines on both task and class-incremental settings with respect to metrics such as accuracy and forgetting.
comment: 30 pages, 7 figures
♻ ☆ Probabilistic Forecasting of Irregular Time Series via Conditional Flows
Probabilistic forecasting of irregularly sampled multivariate time series with missing values is an important problem in many fields, including health care, astronomy, and climate. State-of-the-art methods for the task estimate only marginal distributions of observations in single channels and at single timepoints, assuming a fixed-shape parametric distribution. In this work, we propose a novel model, ProFITi, for probabilistic forecasting of irregularly sampled time series with missing values using conditional normalizing flows. The model learns joint distributions over the future values of the time series conditioned on past observations and queried channels and times, without assuming any fixed shape of the underlying distribution. As model components, we introduce a novel invertible triangular attention layer and an invertible non-linear activation function on and onto the whole real line. We conduct extensive experiments on four datasets and demonstrate that the proposed model provides $4$ times higher likelihood over the previously best model.
♻ ☆ Hardware implementation of timely reliable Bayesian decision-making using memristors
Brains perform decision-making by Bayes theorem. The theorem quantifies events as probabilities and, based on probability rules, renders the decisions. Learning from this, Bayes theorem can be applied to enable efficient user-scene interactions. However, given the probabilistic nature, implementing Bayes theorem in hardware using conventional deterministic computing can incur excessive computational cost and decision latency. Though challenging, here we present a probabilistic computing approach based on memristors to implement the Bayes theorem. We integrate memristors with Boolean logics and, by exploiting the volatile stochastic switching of the memristors, realise probabilistic logic operations, key for hardware Bayes theorem implementation. To empirically validate the efficacy of the hardware Bayes theorem in user-scene interactions, we develop lightweight Bayesian inference and fusion hardware operators using the probabilistic logics and apply the operators in road scene parsing for self-driving, including route planning and obstacle detection. The results show our operators can achieve reliable decisions in less than 0.4 ms (or equivalently 2,500 fps), outperforming human decision-making and the existing driving assistance systems.
♻ ☆ A Spatio-Temporal Neural Network Forecasting Approach for Emulation of Firefront Models
Computational simulations of wildfire spread typically employ empirical rate-of-spread calculations under various conditions (such as terrain, fuel type, weather). Small perturbations in conditions can often lead to significant changes in fire spread (such as speed and direction), necessitating a computationally expensive large set of simulations to quantify uncertainty. Model emulation seeks alternative representations of physical models using machine learning, aiming to provide more efficient and/or simplified surrogate models. We propose a dedicated spatio-temporal neural network based framework for model emulation, able to capture the complex behaviour of fire spread models. The proposed approach can approximate forecasts at fine spatial and temporal resolutions that are often challenging for neural network based approaches. Furthermore, the proposed approach is robust even with small training sets, due to novel data augmentation methods. Empirical experiments show good agreement between simulated and emulated firefronts, with an average Jaccard score of 0.76.
♻ ☆ Learning Spectral Methods by Transformers
Transformers demonstrate significant advantages as the building block of modern LLMs. In this work, we study the capacities of Transformers in performing unsupervised learning. We show that multi-layered Transformers, given a sufficiently large set of pre-training instances, are able to learn the algorithms themselves and perform statistical estimation tasks given new instances. This learning paradigm is distinct from the in-context learning setup and is similar to the learning procedure of human brains where skills are learned through past experience. Theoretically, we prove that pre-trained Transformers can learn the spectral methods and use the classification of bi-class Gaussian mixture model as an example. Our proof is constructive using algorithmic design techniques. Our results are built upon the similarities of multi-layered Transformer architecture with the iterative recovery algorithms used in practice. Empirically, we verify the strong capacity of the multi-layered (pre-trained) Transformer on unsupervised learning through the lens of both the PCA and the Clustering tasks performed on the synthetic and real-world datasets.
comment: 77 pages, 12 figures
♻ ☆ Beyond the Power Law: Estimation, Goodness-of-Fit, and a Semiparametric Extension in Complex Networks
Scale-free networks play a fundamental role in the study of complex networks and various applied fields due to their ability to model a wide range of real-world systems. A key characteristic of these networks is their degree distribution, which often follows a power-law distribution, where the probability mass function is proportional to $x^{-\alpha}$, with $\alpha$ typically ranging between $2 < \alpha < 3$. In this paper, we introduce Bayesian inference methods to obtain more accurate estimates than those obtained using traditional methods, which often yield biased estimates, and precise credible intervals. Through a simulation study, we demonstrate that our approach provides nearly unbiased estimates for the scaling parameter, enhancing the reliability of inferences. We also evaluate new goodness-of-fit tests to improve the effectiveness of the Kolmogorov-Smirnov test, commonly used for this purpose. Our findings show that the Watson test offers superior power while maintaining a controlled type I error rate, enabling us to better determine whether data adheres to a power-law distribution. Finally, we propose a piecewise extension of this model to provide greater flexibility, evaluating the estimation and its goodness-of-fit features as well. In the complex networks field, this extension allows us to model the full degree distribution, instead of just focusing on the tail, as is commonly done. We demonstrate the utility of these novel methods through applications to two real-world datasets, showcasing their practical relevance and potential to advance the analysis of power-law behavior.
comment: 33 pages, 11 figures
♻ ☆ Intelligent System for Automated Molecular Patent Infringement Assessment
Automated drug discovery offers significant potential for accelerating the development of novel therapeutics by substituting labor-intensive human workflows with machine-driven processes. However, molecules generated by artificial intelligence may unintentionally infringe on existing patents, posing legal and financial risks that impede the full automation of drug discovery pipelines. This paper introduces PatentFinder, a novel multi-agent and tool-enhanced intelligence system that can accurately and comprehensively evaluate small molecules for patent infringement. PatentFinder features five specialized agents that collaboratively analyze patent claims and molecular structures with heuristic and model-based tools, generating interpretable infringement reports. To support systematic evaluation, we curate MolPatent-240, a benchmark dataset tailored for patent infringement assessment algorithms. On this benchmark, PatentFinder outperforms baseline methods that rely solely on large language models or specialized chemical tools, achieving a 13.8% improvement in F1-score and a 12% increase in accuracy. Additionally, PatentFinder autonomously generates detailed and interpretable patent infringement reports, showcasing enhanced accuracy and improved interpretability. The high accuracy and interpretability of PatentFinder make it a valuable and reliable tool for automating patent infringement assessments, offering a practical solution for integrating patent protection analysis into the drug discovery pipeline.
♻ ☆ Integral Probability Metrics Meet Neural Networks: The Radon-Kolmogorov-Smirnov Test
Integral probability metrics (IPMs) constitute a general class of nonparametric two-sample tests that are based on maximizing the mean difference between samples from one distribution $P$ versus another $Q$, over all choices of data transformations $f$ living in some function space $\mathcal{F}$. Inspired by recent work that connects what are known as functions of $\textit{Radon bounded variation}$ (RBV) and neural networks (Parhi and Nowak, 2021, 2023), we study the IPM defined by taking $\mathcal{F}$ to be the unit ball in the RBV space of a given smoothness degree $k \geq 0$. This test, which we refer to as the $\textit{Radon-Kolmogorov-Smirnov}$ (RKS) test, can be viewed as a generalization of the well-known and classical Kolmogorov-Smirnov (KS) test to multiple dimensions and higher orders of smoothness. It is also intimately connected to neural networks: we prove that the witness in the RKS test -- the function $f$ achieving the maximum mean difference -- is always a ridge spline of degree $k$, i.e., a single neuron in a neural network. We can thus leverage the power of modern neural network optimization toolkits to (approximately) maximize the criterion that underlies the RKS test. We prove that the RKS test has asymptotically full power at distinguishing any distinct pair $P \not= Q$ of distributions, derive its asymptotic null distribution, and carry out experiments to elucidate the strengths and weaknesses of the RKS test versus the more traditional kernel MMD test.
♻ ☆ The Streetscape Application Services Stack (SASS): Towards a Distributed Sensing Architecture for Urban Applications
As urban populations grow, cities are becoming more complex, driving the deployment of interconnected sensing systems to realize the vision of smart cities. These systems aim to improve safety, mobility, and quality of life through applications that integrate diverse sensors with real-time decision-making. Streetscape applications-focusing on challenges like pedestrian safety and adaptive traffic management-depend on managing distributed, heterogeneous sensor data, aligning information across time and space, and enabling real-time processing. These tasks are inherently complex and often difficult to scale. The Streetscape Application Services Stack (SASS) addresses these challenges with three core services: multimodal data synchronization, spatiotemporal data fusion, and distributed edge computing. By structuring these capabilities as clear, composable abstractions with clear semantics, SASS allows developers to scale streetscape applications efficiently while minimizing the complexity of multimodal integration. We evaluated SASS in two real-world testbed environments: a controlled parking lot and an urban intersection in a major U.S. city. These testbeds allowed us to test SASS under diverse conditions, demonstrating its practical applicability. The Multimodal Data Synchronization service reduced temporal misalignment errors by 88%, achieving synchronization accuracy within 50 milliseconds. Spatiotemporal Data Fusion service improved detection accuracy for pedestrians and vehicles by over 10%, leveraging multicamera integration. The Distributed Edge Computing service increased system throughput by more than an order of magnitude. Together, these results show how SASS provides the abstractions and performance needed to support real-time, scalable urban applications, bridging the gap between sensing infrastructure and actionable streetscape intelligence.
♻ ☆ Stochastic Taylor Derivative Estimator: Efficient amortization for arbitrary differential operators
Optimizing neural networks with loss that contain high-dimensional and high-order differential operators is expensive to evaluate with back-propagation due to $\mathcal{O}(d^{k})$ scaling of the derivative tensor size and the $\mathcal{O}(2^{k-1}L)$ scaling in the computation graph, where $d$ is the dimension of the domain, $L$ is the number of ops in the forward computation graph, and $k$ is the derivative order. In previous works, the polynomial scaling in $d$ was addressed by amortizing the computation over the optimization process via randomization. Separately, the exponential scaling in $k$ for univariate functions ($d=1$) was addressed with high-order auto-differentiation (AD). In this work, we show how to efficiently perform arbitrary contraction of the derivative tensor of arbitrary order for multivariate functions, by properly constructing the input tangents to univariate high-order AD, which can be used to efficiently randomize any differential operator. When applied to Physics-Informed Neural Networks (PINNs), our method provides >1000$\times$ speed-up and >30$\times$ memory reduction over randomization with first-order AD, and we can now solve \emph{1-million-dimensional PDEs in 8 minutes on a single NVIDIA A100 GPU}. This work opens the possibility of using high-order differential operators in large-scale problems.
♻ ☆ Blind Estimation of Sub-band Acoustic Parameters from Ambisonics Recordings using Spectro-Spatial Covariance Features ICASSP2025
Estimating frequency-varying acoustic parameters is essential for enhancing immersive perception in realistic spatial audio creation. In this paper, we propose a unified framework that blindly estimates reverberation time (T60), direct-to-reverberant ratio (DRR), and clarity (C50) across 10 frequency bands using first-order Ambisonics (FOA) speech recordings as inputs. The proposed framework utilizes a novel feature named Spectro-Spatial Covariance Vector (SSCV), efficiently representing temporal, spectral as well as spatial information of the FOA signal. Our models significantly outperform existing single-channel methods with only spectral information, reducing estimation errors by more than half for all three acoustic parameters. Additionally, we introduce FOA-Conv3D, a novel back-end network for effectively utilising the SSCV feature with a 3D convolutional encoder. FOA-Conv3D outperforms the convolutional neural network (CNN) and recurrent convolutional neural network (CRNN) backends, achieving lower estimation errors and accounting for a higher proportion of variance (PoV) for all 3 acoustic parameters.
comment: Accepted by ICASSP2025
♻ ☆ Modeling Latent Neural Dynamics with Gaussian Process Switching Linear Dynamical Systems NeurIPS 2024
Understanding how the collective activity of neural populations relates to computation and ultimately behavior is a key goal in neuroscience. To this end, statistical methods which describe high-dimensional neural time series in terms of low-dimensional latent dynamics have played a fundamental role in characterizing neural systems. Yet, what constitutes a successful method involves two opposing criteria: (1) methods should be expressive enough to capture complex nonlinear dynamics, and (2) they should maintain a notion of interpretability often only warranted by simpler linear models. In this paper, we develop an approach that balances these two objectives: the Gaussian Process Switching Linear Dynamical System (gpSLDS). Our method builds on previous work modeling the latent state evolution via a stochastic differential equation whose nonlinear dynamics are described by a Gaussian process (GP-SDEs). We propose a novel kernel function which enforces smoothly interpolated locally linear dynamics, and therefore expresses flexible -- yet interpretable -- dynamics akin to those of recurrent switching linear dynamical systems (rSLDS). Our approach resolves key limitations of the rSLDS such as artifactual oscillations in dynamics near discrete state boundaries, while also providing posterior uncertainty estimates of the dynamics. To fit our models, we leverage a modified learning objective which improves the estimation accuracy of kernel hyperparameters compared to previous GP-SDE fitting approaches. We apply our method to synthetic data and data recorded in two neuroscience experiments and demonstrate favorable performance in comparison to the rSLDS.
comment: 38th Conference on Neural Information Processing Systems (NeurIPS 2024)
Artificial Intelligence 125
Dataset Distillation via Committee Voting
Dataset distillation aims to synthesize a smaller, representative dataset that preserves the essential properties of the original data, enabling efficient model training with reduced computational resources. Prior work has primarily focused on improving the alignment or matching process between original and synthetic data, or on enhancing the efficiency of distilling large datasets. In this work, we introduce ${\bf C}$ommittee ${\bf V}$oting for ${\bf D}$ataset ${\bf D}$istillation (CV-DD), a novel and orthogonal approach that leverages the collective wisdom of multiple models or experts to create high-quality distilled datasets. We start by showing how to establish a strong baseline that already achieves state-of-the-art accuracy through leveraging recent advancements and thoughtful adjustments in model design and optimization processes. By integrating distributions and predictions from a committee of models while generating high-quality soft labels, our method captures a wider spectrum of data features, reduces model-specific biases and the adverse effects of distribution shifts, leading to significant improvements in generalization. This voting-based strategy not only promotes diversity and robustness within the distilled dataset but also significantly reduces overfitting, resulting in improved performance on post-eval tasks. Extensive experiments across various datasets and IPCs (images per class) demonstrate that Committee Voting leads to more reliable and adaptable distilled data compared to single/multi-model distillation methods, demonstrating its potential for efficient and accurate dataset distillation. Code is available at: https://github.com/Jiacheng8/CV-DD.
comment: Code at: https://github.com/Jiacheng8/CV-DD
☆ UnCommon Objects in 3D
We introduce Uncommon Objects in 3D (uCO3D), a new object-centric dataset for 3D deep learning and 3D generative AI. uCO3D is the largest publicly-available collection of high-resolution videos of objects with 3D annotations that ensures full-360$^{\circ}$ coverage. uCO3D is significantly more diverse than MVImgNet and CO3Dv2, covering more than 1,000 object categories. It is also of higher quality, due to extensive quality checks of both the collected videos and the 3D annotations. Similar to analogous datasets, uCO3D contains annotations for 3D camera poses, depth maps and sparse point clouds. In addition, each object is equipped with a caption and a 3D Gaussian Splat reconstruction. We train several large 3D models on MVImgNet, CO3Dv2, and uCO3D and obtain superior results using the latter, showing that uCO3D is better for learning applications.
☆ WebWalker: Benchmarking LLMs in Web Traversal
Retrieval-augmented generation (RAG) demonstrates remarkable performance across tasks in open-domain question-answering. However, traditional search engines may retrieve shallow content, limiting the ability of LLMs to handle complex, multi-layered information. To address it, we introduce WebWalkerQA, a benchmark designed to assess the ability of LLMs to perform web traversal. It evaluates the capacity of LLMs to traverse a website's subpages to extract high-quality data systematically. We propose WebWalker, which is a multi-agent framework that mimics human-like web navigation through an explore-critic paradigm. Extensive experimental results show that WebWalkerQA is challenging and demonstrates the effectiveness of RAG combined with WebWalker, through the horizontal and vertical integration in real-world scenarios.
☆ Evaluating Agent-based Program Repair at Google
Agent-based program repair offers to automatically resolve complex bugs end-to-end by combining the planning, tool use, and code generation abilities of modern LLMs. Recent work has explored the use of agent-based repair approaches on the popular open-source SWE-Bench, a collection of bugs from highly-rated GitHub Python projects. In addition, various agentic approaches such as SWE-Agent have been proposed to solve bugs in this benchmark. This paper explores the viability of using an agentic approach to address bugs in an enterprise context. To investigate this, we curate an evaluation set of 178 bugs drawn from Google's issue tracking system. This dataset spans both human-reported (78) and machine-reported bugs (100). To establish a repair performance baseline on this benchmark, we implement Passerine, an agent similar in spirit to SWE-Agent that can work within Google's development environment. We show that with 20 trajectory samples and Gemini 1.5 Pro, Passerine can produce a patch that passes bug tests (i.e., plausible) for 73% of machine-reported and 25.6% of human-reported bugs in our evaluation set. After manual examination, we found that 43% of machine-reported bugs and 17.9% of human-reported bugs have at least one patch that is semantically equivalent to the ground-truth patch. These results establish a baseline on an industrially relevant benchmark, which as we show, contains bugs drawn from a different distribution -- in terms of language diversity, size, and spread of changes, etc. -- compared to those in the popular SWE-Bench dataset.
☆ RadAlign: Advancing Radiology Report Generation with Vision-Language Concept Alignment
Automated chest radiographs interpretation requires both accurate disease classification and detailed radiology report generation, presenting a significant challenge in the clinical workflow. Current approaches either focus on classification accuracy at the expense of interpretability or generate detailed but potentially unreliable reports through image captioning techniques. In this study, we present RadAlign, a novel framework that combines the predictive accuracy of vision-language models (VLMs) with the reasoning capabilities of large language models (LLMs). Inspired by the radiologist's workflow, RadAlign first employs a specialized VLM to align visual features with key medical concepts, achieving superior disease classification with an average AUC of 0.885 across multiple diseases. These recognized medical conditions, represented as text-based concepts in the aligned visual-language space, are then used to prompt LLM-based report generation. Enhanced by a retrieval-augmented generation mechanism that grounds outputs in similar historical cases, RadAlign delivers superior report quality with a GREEN score of 0.678, outperforming state-of-the-art methods' 0.634. Our framework maintains strong clinical interpretability while reducing hallucinations, advancing automated medical imaging and report analysis through integrated predictive and generative AI. Code is available at https://github.com/difeigu/RadAlign.
☆ Parallel Key-Value Cache Fusion for Position Invariant RAG
Recent advancements in Large Language Models (LLMs) underscore the necessity of Retrieval Augmented Generation (RAG) to leverage external information. However, LLMs are sensitive to the position of relevant information within contexts and tend to generate incorrect responses when such information is placed in the middle, known as `Lost in the Middle' phenomenon. In this paper, we introduce a framework that generates consistent outputs for decoder-only models, irrespective of the input context order. Experimental results for three open domain question answering tasks demonstrate position invariance, where the model is not sensitive to input context order, and superior robustness to irrelevent passages compared to prevailing approaches for RAG pipelines.
comment: 5 pages
☆ The Paradox of Success in Evolutionary and Bioinspired Optimization: Revisiting Critical Issues, Key Studies, and Methodological Pathways
Evolutionary and bioinspired computation are crucial for efficiently addressing complex optimization problems across diverse application domains. By mimicking processes observed in nature, like evolution itself, these algorithms offer innovative solutions beyond the reach of traditional optimization methods. They excel at finding near-optimal solutions in large, complex search spaces, making them invaluable in numerous fields. However, both areas are plagued by challenges at their core, including inadequate benchmarking, problem-specific overfitting, insufficient theoretical grounding, and superfluous proposals justified only by their biological metaphor. This overview recapitulates and analyzes in depth the criticisms concerning the lack of innovation and rigor in experimental studies within the field. To this end, we examine the judgmental positions of the existing literature in an informed attempt to guide the research community toward directions of solid contribution and advancement in these areas. We summarize guidelines for the design of evolutionary and bioinspired optimizers, the development of experimental comparisons, and the derivation of novel proposals that take a step further in the field. We provide a brief note on automating the process of creating these algorithms, which may help align metaheuristic optimization research with its primary objective (solving real-world problems), provided that our identified pathways are followed. Our conclusions underscore the need for a sustained push towards innovation and the enforcement of methodological rigor in prospective studies to fully realize the potential of these advanced computational techniques.
comment: 38 pages, 1 figure
☆ Inductive Learning of Robot Task Knowledge from Raw Data and Online Expert Feedback
The increasing level of autonomy of robots poses challenges of trust and social acceptance, especially in human-robot interaction scenarios. This requires an interpretable implementation of robotic cognitive capabilities, possibly based on formal methods as logics for the definition of task specifications. However, prior knowledge is often unavailable in complex realistic scenarios. In this paper, we propose an offline algorithm based on inductive logic programming from noisy examples to extract task specifications (i.e., action preconditions, constraints and effects) directly from raw data of few heterogeneous (i.e., not repetitive) robotic executions. Our algorithm leverages on the output of any unsupervised action identification algorithm from video-kinematic recordings. Combining it with the definition of very basic, almost task-agnostic, commonsense concepts about the environment, which contribute to the interpretability of our methodology, we are able to learn logical axioms encoding preconditions of actions, as well as their effects in the event calculus paradigm. Since the quality of learned specifications depends mainly on the accuracy of the action identification algorithm, we also propose an online framework for incremental refinement of task knowledge from user feedback, guaranteeing safe execution. Results in a standard manipulation task and benchmark for user training in the safety-critical surgical robotic scenario, show the robustness, data- and time-efficiency of our methodology, with promising results towards the scalability in more complex domains.
☆ RbRL2.0: Integrated Reward and Policy Learning for Rating-based Reinforcement Learning AAAI 2025
Reinforcement learning (RL), a common tool in decision making, learns policies from various experiences based on the associated cumulative return/rewards without treating them differently. On the contrary, humans often learn to distinguish from different levels of performance and extract the underlying trends towards improving their decision making for best performance. Motivated by this, this paper proposes a novel RL method that mimics humans' decision making process by differentiating among collected experiences for effective policy learning. The main idea is to extract important directional information from experiences with different performance levels, named ratings, so that policies can be updated towards desired deviation from these experiences with different ratings. Specifically, we propose a new policy loss function that penalizes distribution similarities between the current policy and failed experiences with different ratings, and assign different weights to the penalty terms based on the rating classes. Meanwhile, reward learning from these rated samples can be integrated with the new policy loss towards an integrated reward and policy learning from rated samples. Optimizing the integrated reward and policy loss function will lead to the discovery of directions for policy improvement towards maximizing cumulative rewards and penalizing most from the lowest performance level while least from the highest performance level. To evaluate the effectiveness of the proposed method, we present results for experiments on a few typical environments that show improved convergence and overall performance over the existing rating-based reinforcement learning method with only reward learning.
comment: Accepted to the Collaborative AI and Modeling of Humans Bridge Program at AAAI 2025
☆ Data and System Perspectives of Sustainable Artificial Intelligence
Sustainable AI is a subfield of AI for concerning developing and using AI systems in ways of aiming to reduce environmental impact and achieve sustainability. Sustainable AI is increasingly important given that training of and inference with AI models such as large langrage models are consuming a large amount of computing power. In this article, we discuss current issues, opportunities and example solutions for addressing these issues, and future challenges to tackle, from the data and system perspectives, related to data acquisition, data processing, and AI model training and inference.
☆ Smart Learning in the 21st Century: Advancing Constructionism Across Three Digital Epochs
This article explores the evolution of constructionism as an educational framework, tracing its relevance and transformation across three pivotal eras: the advent of personal computing, the networked society, and the current era of generative AI. Rooted in Seymour Papert constructionist philosophy, this study examines how constructionist principles align with the expanding role of digital technology in personal and collective learning. We discuss the transformation of educational environments from hierarchical instructionism to constructionist models that emphasize learner autonomy and interactive, creative engagement. Central to this analysis is the concept of an expanded personality, wherein digital tools and AI integration fundamentally reshape individual self-perception and social interactions. By integrating constructionism into the paradigm of smart education, we propose it as a foundational approach to personalized and democratized learning. Our findings underscore constructionism enduring relevance in navigating the complexities of technology-driven education, providing insights for educators and policymakers seeking to harness digital innovations to foster adaptive, student-centered learning experiences.
comment: 22 pages
☆ TiEBe: A Benchmark for Assessing the Current Knowledge of Large Language Models
In a rapidly evolving knowledge landscape and the increasing adoption of large language models, a need has emerged to keep these models continuously updated with current events. While existing benchmarks evaluate general factual recall, they often overlook two critical aspects: the ability of models to integrate evolving knowledge through continual learning and the significant regional disparities in their performance. To address these gaps, we introduce the Timely Events Benchmark (TiEBe), a dataset containing over 11,000 question-answer pairs focused on globally and regionally significant events. TiEBe leverages structured retrospective data from Wikipedia, enabling continuous updates to assess LLMs' knowledge of evolving global affairs and their understanding of events across different regions. Our benchmark demonstrates that LLMs exhibit substantial geographic disparities in factual recall, emphasizing the need for more balanced global knowledge representation. Furthermore, TiEBe serves as a tool for evaluating continual learning strategies, providing insights into models' ability to acquire new information without forgetting past knowledge.
☆ Estimating Musical Surprisal in Audio ICASSP 2025
In modeling musical surprisal expectancy with computational methods, it has been proposed to use the information content (IC) of one-step predictions from an autoregressive model as a proxy for surprisal in symbolic music. With an appropriately chosen model, the IC of musical events has been shown to correlate with human perception of surprise and complexity aspects, including tonal and rhythmic complexity. This work investigates whether an analogous methodology can be applied to music audio. We train an autoregressive Transformer model to predict compressed latent audio representations of a pretrained autoencoder network. We verify learning effects by estimating the decrease in IC with repetitions. We investigate the mean IC of musical segment types (e.g., A or B) and find that segment types appearing later in a piece have a higher IC than earlier ones on average. We investigate the IC's relation to audio and musical features and find it correlated with timbral variations and loudness and, to a lesser extent, dissonance, rhythmic complexity, and onset density related to audio and musical features. Finally, we investigate if the IC can predict EEG responses to songs and thus model humans' surprisal in music. We provide code for our method on github.com/sonycslparis/audioic.
comment: 5 pages, 2 figures, 1 table. Accepted at the 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2025), Hyderabad, India
☆ A Survey of Embodied AI in Healthcare: Techniques, Applications, and Opportunities
Healthcare systems worldwide face persistent challenges in efficiency, accessibility, and personalization. Powered by modern AI technologies such as multimodal large language models and world models, Embodied AI (EmAI) represents a transformative frontier, offering enhanced autonomy and the ability to interact with the physical world to address these challenges. As an interdisciplinary and rapidly evolving research domain, "EmAI in healthcare" spans diverse fields such as algorithms, robotics, and biomedicine. This complexity underscores the importance of timely reviews and analyses to track advancements, address challenges, and foster cross-disciplinary collaboration. In this paper, we provide a comprehensive overview of the "brain" of EmAI for healthcare, wherein we introduce foundational AI algorithms for perception, actuation, planning, and memory, and focus on presenting the healthcare applications spanning clinical interventions, daily care & companionship, infrastructure support, and biomedical research. Despite its promise, the development of EmAI for healthcare is hindered by critical challenges such as safety concerns, gaps between simulation platforms and real-world applications, the absence of standardized benchmarks, and uneven progress across interdisciplinary domains. We discuss the technical barriers and explore ethical considerations, offering a forward-looking perspective on the future of EmAI in healthcare. A hierarchical framework of intelligent levels for EmAI systems is also introduced to guide further development. By providing systematic insights, this work aims to inspire innovation and practical applications, paving the way for a new era of intelligent, patient-centered healthcare.
comment: 44 pages, 11 figures
☆ Understanding and Benchmarking Artificial Intelligence: OpenAI's o3 Is Not AGI
OpenAI's o3 achieves a high score of 87.5 % on ARC-AGI, a benchmark proposed to measure intelligence. This raises the question whether systems based on Large Language Models (LLMs), particularly o3, demonstrate intelligence and progress towards artificial general intelligence (AGI). Building on the distinction between skills and intelligence made by Fran\c{c}ois Chollet, the creator of ARC-AGI, a new understanding of intelligence is introduced: an agent is the more intelligent, the more efficiently it can achieve the more diverse goals in the more diverse worlds with the less knowledge. An analysis of the ARC-AGI benchmark shows that its tasks represent a very specific type of problem that can be solved by massive trialling of combinations of predefined operations. This method is also applied by o3, achieving its high score through the extensive use of computing power. However, for most problems in the physical world and in the human domain, solutions cannot be tested in advance and predefined operations are not available. Consequently, massive trialling of predefined operations, as o3 does, cannot be a basis for AGI - instead, new approaches are required that can reliably solve a wide variety of problems without existing skills. To support this development, a new benchmark for intelligence is outlined that covers a much higher diversity of unknown tasks to be solved, thus enabling a comprehensive assessment of intelligence and of progress towards AGI.
comment: 15 pages
☆ Online inductive learning from answer sets for efficient reinforcement learning exploration
This paper presents a novel approach combining inductive logic programming with reinforcement learning to improve training performance and explainability. We exploit inductive learning of answer set programs from noisy examples to learn a set of logical rules representing an explainable approximation of the agent policy at each batch of experience. We then perform answer set reasoning on the learned rules to guide the exploration of the learning agent at the next batch, without requiring inefficient reward shaping and preserving optimality with soft bias. The entire procedure is conducted during the online execution of the reinforcement learning algorithm. We preliminarily validate the efficacy of our approach by integrating it into the Q-learning algorithm for the Pac-Man scenario in two maps of increasing complexity. Our methodology produces a significant boost in the discounted return achieved by the agent, even in the first batches of training. Moreover, inductive learning does not compromise the computational time required by Q-learning and learned rules quickly converge to an explanation of the agent policy.
☆ Attention when you need
Being attentive to task-relevant features can improve task performance, but paying attention comes with its own metabolic cost. Therefore, strategic allocation of attention is crucial in performing the task efficiently. This work aims to understand this strategy. Recently, de Gee et al. conducted experiments involving mice performing an auditory sustained attention-value task. This task required the mice to exert attention to identify whether a high-order acoustic feature was present amid the noise. By varying the trial duration and reward magnitude, the task allows us to investigate how an agent should strategically deploy their attention to maximize their benefits and minimize their costs. In our work, we develop a reinforcement learning-based normative model of the mice to understand how it balances attention cost against its benefits. The model is such that at each moment the mice can choose between two levels of attention and decide when to take costly actions that could obtain rewards. Our model suggests that efficient use of attentional resources involves alternating blocks of high attention with blocks of low attention. In the extreme case where the agent disregards sensory input during low attention states, we see that high attention is used rhythmically. Our model provides evidence about how one should deploy attention as a function of task utility, signal statistics, and how attention affects sensory evidence.
☆ Empirical Evaluation of the Implicit Hitting Set Approach for Weighted CSPs
SAT technology has proven to be surprisingly effective in a large variety of domains. However, for the Weighted CSP problem dedicated algorithms have always been superior. One approach not well-studied so far is the use of SAT in conjunction with the Implicit Hitting Set approach. In this work, we explore some alternatives to the existing algorithm of reference. The alternatives, mostly borrowed from related boolean frameworks, consider trade-offs for the two main components of the IHS approach: the computation of low-cost hitting vectors, and their transformation into high-cost cores. For each one, we propose 4 levels of intensity. Since we also test the usefulness of cost function merging, our experiments consider 32 different implementations. Our empirical study shows that for WCSP it is not easy to identify the best alternative. Nevertheless, the cost-function merging encoding and extracting maximal cores seems to be a robust approach.
☆ Diff-Ensembler: Learning to Ensemble 2D Diffusion Models for Volume-to-Volume Medical Image Translation
Despite success in volume-to-volume translations in medical images, most existing models struggle to effectively capture the inherent volumetric distribution using 3D representations. The current state-of-the-art approach combines multiple 2D-based networks through weighted averaging, thereby neglecting the 3D spatial structures. Directly training 3D models in medical imaging presents significant challenges due to high computational demands and the need for large-scale datasets. To address these challenges, we introduce Diff-Ensembler, a novel hybrid 2D-3D model for efficient and effective volumetric translations by ensembling perpendicularly trained 2D diffusion models with a 3D network in each diffusion step. Moreover, our model can naturally be used to ensemble diffusion models conditioned on different modalities, allowing flexible and accurate fusion of input conditions. Extensive experiments demonstrate that Diff-Ensembler attains superior accuracy and volumetric realism in 3D medical image super-resolution and modality translation. We further demonstrate the strength of our model's volumetric realism using tumor segmentation as a downstream task.
☆ An Investigation into Seasonal Variations in Energy Forecasting for Student Residences
This research provides an in-depth evaluation of various machine learning models for energy forecasting, focusing on the unique challenges of seasonal variations in student residential settings. The study assesses the performance of baseline models, such as LSTM and GRU, alongside state-of-the-art forecasting methods, including Autoregressive Feedforward Neural Networks, Transformers, and hybrid approaches. Special attention is given to predicting energy consumption amidst challenges like seasonal patterns, vacations, meteorological changes, and irregular human activities that cause sudden fluctuations in usage. The findings reveal that no single model consistently outperforms others across all seasons, emphasizing the need for season-specific model selection or tailored designs. Notably, the proposed Hyper Network based LSTM and MiniAutoEncXGBoost models exhibit strong adaptability to seasonal variations, effectively capturing abrupt changes in energy consumption during summer months. This study advances the energy forecasting field by emphasizing the critical role of seasonal dynamics and model-specific behavior in achieving accurate predictions.
☆ Initial Findings on Sensor based Open Vocabulary Activity Recognition via Text Embedding Inversion
Conventional human activity recognition (HAR) relies on classifiers trained to predict discrete activity classes, inherently limiting recognition to activities explicitly present in the training set. Such classifiers would invariably fail, putting zero likelihood, when encountering unseen activities. We propose Open Vocabulary HAR (OV-HAR), a framework that overcomes this limitation by first converting each activity into natural language and breaking it into a sequence of elementary motions. This descriptive text is then encoded into a fixed-size embedding. The model is trained to regress this embedding, which is subsequently decoded back into natural language using a pre-trained embedding inversion model. Unlike other works that rely on auto-regressive large language models (LLMs) at their core, OV-HAR achieves open vocabulary recognition without the computational overhead of such models. The generated text can be transformed into a single activity class using LLM prompt engineering. We have evaluated our approach on different modalities, including vision (pose), IMU, and pressure sensors, demonstrating robust generalization across unseen activities and modalities, offering a fundamentally different paradigm from contemporary classifiers.
☆ PROTECT: Protein circadian time prediction using unsupervised learning
Circadian rhythms regulate the physiology and behavior of humans and animals. Despite advancements in understanding these rhythms and predicting circadian phases at the transcriptional level, predicting circadian phases from proteomic data remains elusive. This challenge is largely due to the scarcity of time labels in proteomic datasets, which are often characterized by small sample sizes, high dimensionality, and significant noise. Furthermore, existing methods for predicting circadian phases from transcriptomic data typically rely on prior knowledge of known rhythmic genes, making them unsuitable for proteomic datasets. To address this gap, we developed a novel computational method using unsupervised deep learning techniques to predict circadian sample phases from proteomic data without requiring time labels or prior knowledge of proteins or genes. Our model involves a two-stage training process optimized for robust circadian phase prediction: an initial greedy one-layer-at-a-time pre-training which generates informative initial parameters followed by fine-tuning. During fine-tuning, a specialized loss function guides the model to align protein expression levels with circadian patterns, enabling it to accurately capture the underlying rhythmic structure within the data. We tested our method on both time-labeled and unlabeled proteomic data. For labeled data, we compared our predictions to the known time labels, achieving high accuracy, while for unlabeled human datasets, including postmortem brain regions and urine samples, we explored circadian disruptions. Notably, our analysis identified disruptions in rhythmic proteins between Alzheimer's disease and control subjects across these samples.
☆ Derivation of effective gradient flow equations and dynamical truncation of training data in Deep Learning
We derive explicit equations governing the cumulative biases and weights in Deep Learning with ReLU activation function, based on gradient descent for the Euclidean cost in the input layer, and under the assumption that the weights are, in a precise sense, adapted to the coordinate system distinguished by the activations. We show that gradient descent corresponds to a dynamical process in the input layer, whereby clusters of data are progressively reduced in complexity ("truncated") at an exponential rate that increases with the number of data points that have already been truncated. We provide a detailed discussion of several types of solutions to the gradient flow equations. A main motivation for this work is to shed light on the interpretability question in supervised learning.
comment: AMS Latex, 35 pages
☆ The Essentials of AI for Life and Society: An AI Literacy Course for the University Community AAAI-25
We describe the development of a one-credit course to promote AI literacy at The University of Texas at Austin. In response to a call for the rapid deployment of class to serve a broad audience in Fall of 2023, we designed a 14-week seminar-style course that incorporated an interdisciplinary group of speakers who lectured on topics ranging from the fundamentals of AI to societal concerns including disinformation and employment. University students, faculty, and staff, and even community members outside of the University, were invited to enroll in this online offering: The Essentials of AI for Life and Society. We collected feedback from course participants through weekly reflections and a final survey. Satisfyingly, we found that attendees reported gains in their AI literacy. We sought critical feedback through quantitative and qualitative analysis, which uncovered challenges in designing a course for this general audience. We utilized the course feedback to design a three-credit version of the course that is being offered in Fall of 2024. The lessons we learned and our plans for this new iteration may serve as a guide to instructors designing AI courses for a broad audience.
comment: Accepted to EAAI-25: The 15th Symposium on Educational Advances in Artificial Intelligence, collocated with AAAI-25
☆ Enhancing Retrieval-Augmented Generation: A Study of Best Practices
Retrieval-Augmented Generation (RAG) systems have recently shown remarkable advancements by integrating retrieval mechanisms into language models, enhancing their ability to produce more accurate and contextually relevant responses. However, the influence of various components and configurations within RAG systems remains underexplored. A comprehensive understanding of these elements is essential for tailoring RAG systems to complex retrieval tasks and ensuring optimal performance across diverse applications. In this paper, we develop several advanced RAG system designs that incorporate query expansion, various novel retrieval strategies, and a novel Contrastive In-Context Learning RAG. Our study systematically investigates key factors, including language model size, prompt design, document chunk size, knowledge base size, retrieval stride, query expansion techniques, Contrastive In-Context Learning knowledge bases, multilingual knowledge bases, and Focus Mode retrieving relevant context at sentence-level. Through extensive experimentation, we provide a detailed analysis of how these factors influence response quality. Our findings offer actionable insights for developing RAG systems, striking a balance between contextual richness and retrieval-generation efficiency, thereby paving the way for more adaptable and high-performing RAG frameworks in diverse real-world scenarios. Our code and implementation details are publicly available.
☆ Information-Theoretic Dual Memory System for Continual Learning
Continuously acquiring new knowledge from a dynamic environment is a fundamental capability for animals, facilitating their survival and ability to address various challenges. This capability is referred to as continual learning, which focuses on the ability to learn a sequence of tasks without the detriment of previous knowledge. A prevalent strategy to tackle continual learning involves selecting and storing numerous essential data samples from prior tasks within a fixed-size memory buffer. However, the majority of current memory-based techniques typically utilize a single memory buffer, which poses challenges in concurrently managing newly acquired and previously learned samples. Drawing inspiration from the Complementary Learning Systems (CLS) theory, which defines rapid and gradual learning mechanisms for processing information, we propose an innovative dual memory system called the Information-Theoretic Dual Memory System (ITDMS). This system comprises a fast memory buffer designed to retain temporary and novel samples, alongside a slow memory buffer dedicated to preserving critical and informative samples. The fast memory buffer is optimized employing an efficient reservoir sampling process. Furthermore, we introduce a novel information-theoretic memory optimization strategy that selectively identifies and retains diverse and informative data samples for the slow memory buffer. Additionally, we propose a novel balanced sample selection procedure that automatically identifies and eliminates redundant memorized samples, thus freeing up memory capacity for new data acquisitions, which can deal with a growing array of tasks. Our methodology is rigorously assessed through a series of continual learning experiments, with empirical results underscoring the effectiveness of the proposed system.
comment: 35 pages, 9 figures, submitted to Knowledge-Based Systems
☆ Emergent effects of scaling on the functional hierarchies within large language models
Large language model (LLM) architectures are often described as functionally hierarchical: Early layers process syntax, middle layers begin to parse semantics, and late layers integrate information. The present work revisits these ideas. This research submits simple texts to an LLM (e.g., "A church and organ") and extracts the resulting activations. Then, for each layer, support vector machines and ridge regressions are fit to predict a text's label and thus examine whether a given layer encodes some information. Analyses using a small model (Llama-3.2-3b; 28 layers) partly bolster the common hierarchical perspective: Item-level semantics are most strongly represented early (layers 2-7), then two-item relations (layers 8-12), and then four-item analogies (layers 10-15). Afterward, the representation of items and simple relations gradually decreases in deeper layers that focus on more global information. However, several findings run counter to a steady hierarchy view: First, although deep layers can represent document-wide abstractions, deep layers also compress information from early portions of the context window without meaningful abstraction. Second, when examining a larger model (Llama-3.3-70b-Instruct), stark fluctuations in abstraction level appear: As depth increases, two-item relations and four-item analogies initially increase in their representation, then markedly decrease, and afterward increase again momentarily. This peculiar pattern consistently emerges across several experiments. Third, another emergent effect of scaling is coordination between the attention mechanisms of adjacent layers. Across multiple experiments using the larger model, adjacent layers fluctuate between what information they each specialize in representing. In sum, an abstraction hierarchy often manifests across layers, but large models also deviate from this structure in curious ways.
☆ TempoGPT: Enhancing Temporal Reasoning via Quantizing Embedding
Multi-modal language model has made advanced progress in vision and audio, but still faces significant challenges in dealing with complex reasoning tasks in the time series domain. The reasons are twofold. First, labels for multi-modal time series data are coarse and devoid of analysis or reasoning processes. Training with these data cannot improve the model's reasoning capabilities. Second, due to the lack of precise tokenization in processing time series, the representation patterns for temporal and textual information are inconsistent, which hampers the effectiveness of multi-modal alignment. To address these challenges, we propose a multi-modal time series data construction approach and a multi-modal time series language model (TLM), TempoGPT. Specially, we construct multi-modal data for complex reasoning tasks by analyzing the variable-system relationships within a white-box system. Additionally, proposed TempoGPT achieves consistent representation between temporal and textual information by quantizing temporal embeddings, where temporal embeddings are quantized into a series of discrete tokens using a predefined codebook; subsequently, a shared embedding layer processes both temporal and textual tokens. Extensive experiments demonstrate that TempoGPT accurately perceives temporal information, logically infers conclusions, and achieves state-of-the-art in the constructed complex time series reasoning tasks. Moreover, we quantitatively demonstrate the effectiveness of quantizing temporal embeddings in enhancing multi-modal alignment and the reasoning capabilities of TLMs. Code and data are available at https://github.com/zhanghaochuan20/TempoGPT.
☆ Anonymization of Documents for Law Enforcement with Machine Learning
The steadily increasing utilization of data-driven methods and approaches in areas that handle sensitive personal information such as in law enforcement mandates an ever increasing effort in these institutions to comply with data protection guidelines. In this work, we present a system for automatically anonymizing images of scanned documents, reducing manual effort while ensuring data protection compliance. Our method considers the viability of further forensic processing after anonymization by minimizing automatically redacted areas by combining automatic detection of sensitive regions with knowledge from a manually anonymized reference document. Using a self-supervised image model for instance retrieval of the reference document, our approach requires only one anonymized example to efficiently redact all documents of the same type, significantly reducing processing time. We show that our approach outperforms both a purely automatic redaction system and also a naive copy-paste scheme of the reference anonymization to other documents on a hand-crafted dataset of ground truth redactions.
comment: Accepted at IEEE Symposium on CI in Security, Defence and Biometrics 2025 (IEEE CISDB)
☆ Evaluation of Artificial Intelligence Methods for Lead Time Prediction in Non-Cycled Areas of Automotive Production
The present study examines the effectiveness of applying Artificial Intelligence methods in an automotive production environment to predict unknown lead times in a non-cycle-controlled production area. Data structures are analyzed to identify contextual features and then preprocessed using one-hot encoding. Methods selection focuses on supervised machine learning techniques. In supervised learning methods, regression and classification methods are evaluated. Continuous regression based on target size distribution is not feasible. Classification methods analysis shows that Ensemble Learning and Support Vector Machines are the most suitable. Preliminary study results indicate that gradient boosting algorithms LightGBM, XGBoost, and CatBoost yield the best results. After further testing and extensive hyperparameter optimization, the final method choice is the LightGBM algorithm. Depending on feature availability and prediction interval granularity, relative prediction accuracies of up to 90% can be achieved. Further tests highlight the importance of periodic retraining of AI models to accurately represent complex production processes using the database. The research demonstrates that AI methods can be effectively applied to highly variable production data, adding business value by providing an additional metric for various control tasks while outperforming current non AI-based systems.
comment: 7 pages, 4 figures, CLC2024 Conference
☆ The Lessons of Developing Process Reward Models in Mathematical Reasoning
Process Reward Models (PRMs) emerge as a promising approach for process supervision in mathematical reasoning of Large Language Models (LLMs), which aim to identify and mitigate intermediate errors in the reasoning processes. However, the development of effective PRMs faces significant challenges, particularly in data annotation and evaluation methodologies. In this paper, through extensive experiments, we demonstrate that commonly used Monte Carlo (MC) estimation-based data synthesis for PRMs typically yields inferior performance and generalization compared to LLM-as-a-judge and human annotation methods. MC estimation relies on completion models to evaluate current-step correctness, leading to inaccurate step verification. Furthermore, we identify potential biases in conventional Best-of-N (BoN) evaluation strategies for PRMs: (1) The unreliable policy models generate responses with correct answers but flawed processes, leading to a misalignment between the evaluation criteria of BoN and the PRM objectives of process verification. (2) The tolerance of PRMs of such responses leads to inflated BoN scores. (3) Existing PRMs have a significant proportion of minimum scores concentrated on the final answer steps, revealing the shift from process to outcome-based assessment in BoN Optimized PRMs. To address these challenges, we develop a consensus filtering mechanism that effectively integrates MC estimation with LLM-as-a-judge and advocates a more comprehensive evaluation framework that combines response-level and step-level metrics. Based on the mechanisms, we significantly improve both model performance and data efficiency in the BoN evaluation and the step-wise error identification task. Finally, we release a new state-of-the-art PRM that outperforms existing open-source alternatives and provides practical guidelines for future research in building process supervision models.
☆ Principles for Responsible AI Consciousness Research
Recent research suggests that it may be possible to build conscious AI systems now or in the near future. Conscious AI systems would arguably deserve moral consideration, and it may be the case that large numbers of conscious systems could be created and caused to suffer. Furthermore, AI systems or AI-generated characters may increasingly give the impression of being conscious, leading to debate about their moral status. Organisations involved in AI research must establish principles and policies to guide research and deployment choices and public communication concerning consciousness. Even if an organisation chooses not to study AI consciousness as such, it will still need policies in place, as those developing advanced AI systems risk inadvertently creating conscious entities. Responsible research and deployment practices are essential to address this possibility. We propose five principles for responsible research and argue that research organisations should make voluntary, public commitments to principles on these lines. Our principles concern research objectives and procedures, knowledge sharing and public communications.
☆ LLM-Net: Democratizing LLMs-as-a-Service through Blockchain-based Expert Networks
The centralization of Large Language Models (LLMs) development has created significant barriers to AI advancement, limiting the democratization of these powerful technologies. This centralization, coupled with the scarcity of high-quality training data and mounting complexity of maintaining comprehensive expertise across rapidly expanding knowledge domains, poses critical challenges to the continued growth of LLMs. While solutions like Retrieval-Augmented Generation (RAG) offer potential remedies, maintaining up-to-date expert knowledge across diverse domains remains a significant challenge, particularly given the exponential growth of specialized information. This paper introduces LLMs Networks (LLM-Net), a blockchain-based framework that democratizes LLMs-as-a-Service through a decentralized network of specialized LLM providers. By leveraging collective computational resources and distributed domain expertise, LLM-Net incorporates fine-tuned expert models for various specific domains, ensuring sustained knowledge growth while maintaining service quality through collaborative prompting mechanisms. The framework's robust design includes blockchain technology for transparent transaction and performance validation, establishing an immutable record of service delivery. Our simulation, built on top of state-of-the-art LLMs such as Claude 3.5 Sonnet, Llama 3.1, Grok-2, and GPT-4o, validates the effectiveness of the reputation-based mechanism in maintaining service quality by selecting high-performing respondents (LLM providers). Thereby it demonstrates the potential of LLM-Net to sustain AI advancement through the integration of decentralized expertise and blockchain-based accountability.
☆ Lifelong Learning of Large Language Model based Agents: A Roadmap
Lifelong learning, also known as continual or incremental learning, is a crucial component for advancing Artificial General Intelligence (AGI) by enabling systems to continuously adapt in dynamic environments. While large language models (LLMs) have demonstrated impressive capabilities in natural language processing, existing LLM agents are typically designed for static systems and lack the ability to adapt over time in response to new challenges. This survey is the first to systematically summarize the potential techniques for incorporating lifelong learning into LLM-based agents. We categorize the core components of these agents into three modules: the perception module for multimodal input integration, the memory module for storing and retrieving evolving knowledge, and the action module for grounded interactions with the dynamic environment. We highlight how these pillars collectively enable continuous adaptation, mitigate catastrophic forgetting, and improve long-term performance. This survey provides a roadmap for researchers and practitioners working to develop lifelong learning capabilities in LLM agents, offering insights into emerging trends, evaluation metrics, and application scenarios. Relevant literature and resources are available at \href{this url}{https://github.com/qianlima-lab/awesome-lifelong-llm-agent}.
comment: 46 pages
☆ Bridging Smart Meter Gaps: A Benchmark of Statistical, Machine Learning and Time Series Foundation Models for Data Imputation
The integrity of time series data in smart grids is often compromised by missing values due to sensor failures, transmission errors, or disruptions. Gaps in smart meter data can bias consumption analyses and hinder reliable predictions, causing technical and economic inefficiencies. As smart meter data grows in volume and complexity, conventional techniques struggle with its nonlinear and nonstationary patterns. In this context, Generative Artificial Intelligence offers promising solutions that may outperform traditional statistical methods. In this paper, we evaluate two general-purpose Large Language Models and five Time Series Foundation Models for smart meter data imputation, comparing them with conventional Machine Learning and statistical models. We introduce artificial gaps (30 minutes to one day) into an anonymized public dataset to test inference capabilities. Results show that Time Series Foundation Models, with their contextual understanding and pattern recognition, could significantly enhance imputation accuracy in certain cases. However, the trade-off between computational cost and performance gains remains a critical consideration.
☆ Skip Mamba Diffusion for Monocular 3D Semantic Scene Completion AAAI 2025
3D semantic scene completion is critical for multiple downstream tasks in autonomous systems. It estimates missing geometric and semantic information in the acquired scene data. Due to the challenging real-world conditions, this task usually demands complex models that process multi-modal data to achieve acceptable performance. We propose a unique neural model, leveraging advances from the state space and diffusion generative modeling to achieve remarkable 3D semantic scene completion performance with monocular image input. Our technique processes the data in the conditioned latent space of a variational autoencoder where diffusion modeling is carried out with an innovative state space technique. A key component of our neural network is the proposed Skimba (Skip Mamba) denoiser, which is adept at efficiently processing long-sequence data. The Skimba diffusion model is integral to our 3D scene completion network, incorporating a triple Mamba structure, dimensional decomposition residuals and varying dilations along three directions. We also adopt a variant of this network for the subsequent semantic segmentation stage of our method. Extensive evaluation on the standard SemanticKITTI and SSCBench-KITTI360 datasets show that our approach not only outperforms other monocular techniques by a large margin, it also achieves competitive performance against stereo methods. The code is available at https://github.com/xrkong/skimba
comment: Accepted by AAAI 2025
☆ MOS-Attack: A Scalable Multi-objective Adversarial Attack Framework CVPR 2025
Crafting adversarial examples is crucial for evaluating and enhancing the robustness of Deep Neural Networks (DNNs), presenting a challenge equivalent to maximizing a non-differentiable 0-1 loss function. However, existing single objective methods, namely adversarial attacks focus on a surrogate loss function, do not fully harness the benefits of engaging multiple loss functions, as a result of insufficient understanding of their synergistic and conflicting nature. To overcome these limitations, we propose the Multi-Objective Set-based Attack (MOS Attack), a novel adversarial attack framework leveraging multiple loss functions and automatically uncovering their interrelations. The MOS Attack adopts a set-based multi-objective optimization strategy, enabling the incorporation of numerous loss functions without additional parameters. It also automatically mines synergistic patterns among various losses, facilitating the generation of potent adversarial attacks with fewer objectives. Extensive experiments have shown that our MOS Attack outperforms single-objective attacks. Furthermore, by harnessing the identified synergistic patterns, MOS Attack continues to show superior results with a reduced number of loss functions.
comment: Under Review of CVPR 2025
☆ Lessons From Red Teaming 100 Generative AI Products
In recent years, AI red teaming has emerged as a practice for probing the safety and security of generative AI systems. Due to the nascency of the field, there are many open questions about how red teaming operations should be conducted. Based on our experience red teaming over 100 generative AI products at Microsoft, we present our internal threat model ontology and eight main lessons we have learned: 1. Understand what the system can do and where it is applied 2. You don't have to compute gradients to break an AI system 3. AI red teaming is not safety benchmarking 4. Automation can help cover more of the risk landscape 5. The human element of AI red teaming is crucial 6. Responsible AI harms are pervasive but difficult to measure 7. LLMs amplify existing security risks and introduce new ones 8. The work of securing AI systems will never be complete By sharing these insights alongside case studies from our operations, we offer practical recommendations aimed at aligning red teaming efforts with real world risks. We also highlight aspects of AI red teaming that we believe are often misunderstood and discuss open questions for the field to consider.
☆ Breaking Memory Limits: Gradient Wavelet Transform Enhances LLMs Training
Large language models (LLMs) have shown impressive performance across a range of natural language processing tasks. However, their vast number of parameters introduces significant memory challenges during training, particularly when using memory-intensive optimizers like Adam. Existing memory-efficient algorithms often rely on techniques such as singular value decomposition projection or weight freezing. While these approaches help alleviate memory constraints, they generally produce suboptimal results compared to full-rank updates. In this paper, we investigate the memory-efficient method beyond low-rank training, proposing a novel solution called Gradient Wavelet Transform (GWT), which applies wavelet transforms to gradients in order to significantly reduce the memory requirements for maintaining optimizer states. We demonstrate that GWT can be seamlessly integrated with memory-intensive optimizers, enabling efficient training without sacrificing performance. Through extensive experiments on both pre-training and fine-tuning tasks, we show that GWT achieves state-of-the-art performance compared with advanced memory-efficient optimizers and full-rank approaches in terms of both memory usage and training performance.
☆ Exploring the Use of Contrastive Language-Image Pre-Training for Human Posture Classification: Insights from Yoga Pose Analysis
Accurate human posture classification in images and videos is crucial for automated applications across various fields, including work safety, physical rehabilitation, sports training, or daily assisted living. Recently, multimodal learning methods, such as Contrastive Language-Image Pretraining (CLIP), have advanced significantly in jointly understanding images and text. This study aims to assess the effectiveness of CLIP in classifying human postures, focusing on its application in yoga. Despite the initial limitations of the zero-shot approach, applying transfer learning on 15,301 images (real and synthetic) with 82 classes has shown promising results. The article describes the full procedure for fine-tuning, including the choice for image description syntax, models and hyperparameters adjustment. The fine-tuned CLIP model, tested on 3826 images, achieves an accuracy of over 85%, surpassing the current state-of-the-art of previous works on the same dataset by approximately 6%, its training time being 3.5 times lower than what is needed to fine-tune a YOLOv8-based model. For more application-oriented scenarios, with smaller datasets of six postures each, containing 1301 and 401 training images, the fine-tuned models attain an accuracy of 98.8% and 99.1%, respectively. Furthermore, our experiments indicate that training with as few as 20 images per pose can yield around 90% accuracy in a six-class dataset. This study demonstrates that this multimodal technique can be effectively used for yoga pose classification, and possibly for human posture classification, in general. Additionally, CLIP inference time (around 7 ms) supports that the model can be integrated into automated systems for posture evaluation, e.g., for developing a real-time personal yoga assistant for performance assessment.
☆ Multi-face emotion detection for effective Human-Robot Interaction
The integration of dialogue interfaces in mobile devices has become ubiquitous, providing a wide array of services. As technology progresses, humanoid robots designed with human-like features to interact effectively with people are gaining prominence, and the use of advanced human-robot dialogue interfaces is continually expanding. In this context, emotion recognition plays a crucial role in enhancing human-robot interaction by enabling robots to understand human intentions. This research proposes a facial emotion detection interface integrated into a mobile humanoid robot, capable of displaying real-time emotions from multiple individuals on a user interface. To this end, various deep neural network models for facial expression recognition were developed and evaluated under consistent computer-based conditions, yielding promising results. Afterwards, a trade-off between accuracy and memory footprint was carefully considered to effectively implement this application on a mobile humanoid robot.
comment: 9 pages, 8 figures and 1 table. Accepted at the 17th International Conference on Agents and Artificial Intelligence (ICAART 2025), Porto, Portugal
☆ Crowdsourced human-based computational approach for tagging peripheral blood smear sample images from Sickle Cell Disease patients using non-expert users
In this paper, we present a human-based computation approach for the analysis of peripheral blood smear (PBS) images images in patients with Sickle Cell Disease (SCD). We used the Mechanical Turk microtask market to crowdsource the labeling of PBS images. We then use the expert-tagged erythrocytesIDB dataset to assess the accuracy and reliability of our proposal. Our results showed that when a robust consensus is achieved among the Mechanical Turk workers, probability of error is very low, based on comparison with expert analysis. This suggests that our proposed approach can be used to annotate datasets of PBS images, which can then be used to train automated methods for the diagnosis of SCD. In future work, we plan to explore the potential integration of our findings with outcomes obtained through automated methodologies. This could lead to the development of more accurate and reliable methods for the diagnosis of SCD
☆ Generalizable Graph Neural Networks for Robust Power Grid Topology Control
The energy transition necessitates new congestion management methods. One such method is controlling the grid topology with machine learning (ML). This approach has gained popularity following the Learning to Run a Power Network (L2RPN) competitions. Graph neural networks (GNNs) are a class of ML models that reflect graph structure in their computation, which makes them suitable for power grid modeling. Various GNN approaches for topology control have thus been proposed. We propose the first GNN model for grid topology control that uses only GNN layers. Additionally, we identify the busbar information asymmetry problem that the popular homogeneous graph representation suffers from, and propose a heterogeneous graph representation to resolve it. We train both homogeneous and heterogeneous GNNs and fully connected neural networks (FCNN) baselines on an imitation learning task. We evaluate the models according to their classification accuracy and grid operation ability. We find that the heterogeneous GNNs perform best on in-distribution networks, followed by the FCNNs, and lastly, the homogeneous GNNs. We also find that both GNN types generalize better to out-of-distribution networks than FCNNs.
☆ Kriging and Gaussian Process Interpolation for Georeferenced Data Augmentation
Data augmentation is a crucial step in the development of robust supervised learning models, especially when dealing with limited datasets. This study explores interpolation techniques for the augmentation of geo-referenced data, with the aim of predicting the presence of Commelina benghalensis L. in sugarcane plots in La R{\'e}union. Given the spatial nature of the data and the high cost of data collection, we evaluated two interpolation approaches: Gaussian processes (GPs) with different kernels and kriging with various variograms. The objectives of this work are threefold: (i) to identify which interpolation methods offer the best predictive performance for various regression algorithms, (ii) to analyze the evolution of performance as a function of the number of observations added, and (iii) to assess the spatial consistency of augmented datasets. The results show that GP-based methods, in particular with combined kernels (GP-COMB), significantly improve the performance of regression algorithms while requiring less additional data. Although kriging shows slightly lower performance, it is distinguished by a more homogeneous spatial coverage, a potential advantage in certain contexts.
☆ The Spoils of Algorithmic Collusion: Profit Allocation Among Asymmetric Firms
We study the propensity of independent algorithms to collude in repeated Cournot duopoly games. Specifically, we investigate the predictive power of different oligopoly and bargaining solutions regarding the effect of asymmetry between firms. We find that both consumers and firms can benefit from asymmetry. Algorithms produce more competitive outcomes when firms are symmetric, but less when they are very asymmetric. Although the static Nash equilibrium underestimates the effect on total quantity and overestimates the effect on profits, it delivers surprisingly accurate predictions in terms of total welfare. The best description of our results is provided by the equal relative gains solution. In particular, we find algorithms to agree on profits that are on or close to the Pareto frontier for all degrees of asymmetry. Our results suggest that the common belief that symmetric industries are more prone to collusion may no longer hold when algorithms increasingly drive managerial decisions.
☆ Anomalous Agreement: How to find the Ideal Number of Anomaly Classes in Correlated, Multivariate Time Series Data AAAI
Detecting and classifying abnormal system states is critical for condition monitoring, but supervised methods often fall short due to the rarity of anomalies and the lack of labeled data. Therefore, clustering is often used to group similar abnormal behavior. However, evaluating cluster quality without ground truth is challenging, as existing measures such as the Silhouette Score (SSC) only evaluate the cohesion and separation of clusters and ignore possible prior knowledge about the data. To address this challenge, we introduce the Synchronized Anomaly Agreement Index (SAAI), which exploits the synchronicity of anomalies across multivariate time series to assess cluster quality. We demonstrate the effectiveness of SAAI by showing that maximizing SAAI improves accuracy on the task of finding the true number of anomaly classes K in correlated time series by 0.23 compared to SSC and by 0.32 compared to X-Means. We also show that clusters obtained by maximizing SAAI are easier to interpret compared to SSC.
comment: Acccepted at AAAI Workshop on AI for Time Series Analysis (AI4TS) 2025
☆ Natural Language-Assisted Multi-modal Medication Recommendation
Combinatorial medication recommendation(CMR) is a fundamental task of healthcare, which offers opportunities for clinical physicians to provide more precise prescriptions for patients with intricate health conditions, particularly in the scenarios of long-term medical care. Previous research efforts have sought to extract meaningful information from electronic health records (EHRs) to facilitate combinatorial medication recommendations. Existing learning-based approaches further consider the chemical structures of medications, but ignore the textual medication descriptions in which the functionalities are clearly described. Furthermore, the textual knowledge derived from the EHRs of patients remains largely underutilized. To address these issues, we introduce the Natural Language-Assisted Multi-modal Medication Recommendation(NLA-MMR), a multi-modal alignment framework designed to learn knowledge from the patient view and medication view jointly. Specifically, NLA-MMR formulates CMR as an alignment problem from patient and medication modalities. In this vein, we employ pretrained language models(PLMs) to extract in-domain knowledge regarding patients and medications, serving as the foundational representation for both modalities. In the medication modality, we exploit both chemical structures and textual descriptions to create medication representations. In the patient modality, we generate the patient representations based on textual descriptions of diagnosis, procedure, and symptom. Extensive experiments conducted on three publicly accessible datasets demonstrate that NLA-MMR achieves new state-of-the-art performance, with a notable average improvement of 4.72% in Jaccard score. Our source code is publicly available on https://github.com/jtan1102/NLA-MMR_CIKM_2024.
comment: 10 pages
☆ QuantuneV2: Compiler-Based Local Metric-Driven Mixed Precision Quantization for Practical Embedded AI Applications
Mixed-precision quantization methods have been proposed to reduce model size while minimizing accuracy degradation. However, existing studies require retraining and do not consider the computational overhead and intermediate representations (IR) generated during the compilation process, limiting their application at the compiler level. This computational overhead refers to the runtime latency caused by frequent quantization and dequantization operations during inference. Performing these operations at the individual operator level causes significant runtime delays. To address these issues, we propose QuantuneV2, a compiler-based mixed-precision quantization method designed for practical embedded AI applications. QuantuneV2 performs inference only twice, once before quantization and once after quantization, and operates with a computational complexity of O(n) that increases linearly with the number of model parameters. We also made the sensitivity analysis more stable by using local metrics like weights, activation values, the Signal to Quantization Noise Ratio, and the Mean Squared Error. We also cut down on computational overhead by choosing the best IR and using operator fusion. Experimental results show that QuantuneV2 achieved up to a 10.28 percent improvement in accuracy and a 12.52 percent increase in speed compared to existing methods across five models: ResNet18v1, ResNet50v1, SqueezeNetv1, VGGNet, and MobileNetv2. This demonstrates that QuantuneV2 enhances model performance while maintaining computational efficiency, making it suitable for deployment in embedded AI environments.
comment: 18 pages, 10 figures, Accepted in Future Generation Computer Systems Journal
☆ Eye Sclera for Fair Face Image Quality Assessment
Fair operational systems are crucial in gaining and maintaining society's trust in face recognition systems (FRS). FRS start with capturing an image and assessing its quality before using it further for enrollment or verification. Fair Face Image Quality Assessment (FIQA) schemes therefore become equally important in the context of fair FRS. This work examines the sclera as a quality assessment region for obtaining a fair FIQA. The sclera region is agnostic to demographic variations and skin colour for assessing the quality of a face image. We analyze three skin tone related ISO/IEC face image quality assessment measures and assess the sclera region as an alternative area for assessing FIQ. Our analysis of the face dataset of individuals from different demographic groups representing different skin tones indicates sclera as an alternative to measure dynamic range, over- and under-exposure of face using sclera region alone. The sclera region being agnostic to skin tone, i.e., demographic factors, provides equal utility as a fair FIQA as shown by our Error-vs-Discard Characteristic (EDC) curve analysis.
☆ CureGraph: Contrastive Multi-Modal Graph Representation Learning for Urban Living Circle Health Profiling and Prediction
The early detection and prediction of health status decline among the elderly at the neighborhood level are of great significance for urban planning and public health policymaking. While existing studies affirm the connection between living environments and health outcomes, most rely on single data modalities or simplistic feature concatenation of multi-modal information, limiting their ability to comprehensively profile the health-oriented urban environments. To fill this gap, we propose CureGraph, a contrastive multi-modal representation learning framework for urban health prediction that employs graph-based techniques to infer the prevalence of common chronic diseases among the elderly within the urban living circles of each neighborhood. CureGraph leverages rich multi-modal information, including photos and textual reviews of residential areas and their surrounding points of interest, to generate urban neighborhood embeddings. By integrating pre-trained visual and textual encoders with graph modeling techniques, CureGraph captures cross-modal spatial dependencies, offering a comprehensive understanding of urban environments tailored to elderly health considerations. Extensive experiments on real-world datasets demonstrate that CureGraph improves the best baseline by $28\%$ on average in terms of $R^2$ across elderly disease risk prediction tasks. Moreover, the model enables the identification of stage-wise chronic disease progression and supports comparative public health analysis across neighborhoods, offering actionable insights for sustainable urban development and enhanced quality of life. The code is publicly available at https://github.com/jinlin2021/CureGraph.
☆ TIMRL: A Novel Meta-Reinforcement Learning Framework for Non-Stationary and Multi-Task Environments
In recent years, meta-reinforcement learning (meta-RL) algorithm has been proposed to improve sample efficiency in the field of decision-making and control, enabling agents to learn new knowledge from a small number of samples. However, most research uses the Gaussian distribution to extract task representation, which is poorly adapted to tasks that change in non-stationary environment. To address this problem, we propose a novel meta-reinforcement learning method by leveraging Gaussian mixture model and the transformer network to construct task inference model. The Gaussian mixture model is utilized to extend the task representation and conduct explicit encoding of tasks. Specifically, the classification of tasks is encoded through transformer network to determine the Gaussian component corresponding to the task. By leveraging task labels, the transformer network is trained using supervised learning. We validate our method on MuJoCo benchmarks with non-stationary and multi-task environments. Experimental results demonstrate that the proposed method dramatically improves sample efficiency and accurately recognizes the classification of the tasks, while performing excellently in the environment.
☆ FlexQuant: Elastic Quantization Framework for Locally Hosted LLM on Edge Devices
Deploying LLMs on edge devices presents serious technical challenges. Memory elasticity is crucial for edge devices with unified memory, where memory is shared and fluctuates dynamically. Existing solutions suffer from either poor transition granularity or high storage costs. We propose FlexQuant, a novel elasticity framework that generates an ensemble of quantized models, providing an elastic hosting solution with 15x granularity improvement and 10x storage reduction compared to SoTA methods. FlexQuant works with most quantization methods and creates a family of trade-off options under various storage limits through our pruning method. It brings great performance and flexibility to the edge deployment of LLMs.
☆ How GPT learns layer by layer
Large Language Models (LLMs) excel at tasks like language processing, strategy games, and reasoning but struggle to build generalizable internal representations essential for adaptive decision-making in agents. For agents to effectively navigate complex environments, they must construct reliable world models. While LLMs perform well on specific benchmarks, they often fail to generalize, leading to brittle representations that limit their real-world effectiveness. Understanding how LLMs build internal world models is key to developing agents capable of consistent, adaptive behavior across tasks. We analyze OthelloGPT, a GPT-based model trained on Othello gameplay, as a controlled testbed for studying representation learning. Despite being trained solely on next-token prediction with random valid moves, OthelloGPT shows meaningful layer-wise progression in understanding board state and gameplay. Early layers capture static attributes like board edges, while deeper layers reflect dynamic tile changes. To interpret these representations, we compare Sparse Autoencoders (SAEs) with linear probes, finding that SAEs offer more robust, disentangled insights into compositional features, whereas linear probes mainly detect features useful for classification. We use SAEs to decode features related to tile color and tile stability, a previously unexamined feature that reflects complex gameplay concepts like board control and long-term planning. We study the progression of linear probe accuracy and tile color using both SAE's and linear probes to compare their effectiveness at capturing what the model is learning. Although we begin with a smaller language model, OthelloGPT, this study establishes a framework for understanding the internal representations learned by GPT models, transformers, and LLMs more broadly. Our code is publicly available: https://github.com/ALT-JS/OthelloSAE.
☆ AdaCS: Adaptive Normalization for Enhanced Code-Switching ASR ICASSP 2025
Intra-sentential code-switching (CS) refers to the alternation between languages that happens within a single utterance and is a significant challenge for Automatic Speech Recognition (ASR) systems. For example, when a Vietnamese speaker uses foreign proper names or specialized terms within their speech. ASR systems often struggle to accurately transcribe intra-sentential CS due to their training on monolingual data and the unpredictable nature of CS. This issue is even more pronounced for low-resource languages, where limited data availability hinders the development of robust models. In this study, we propose AdaCS, a normalization model integrates an adaptive bias attention module (BAM) into encoder-decoder network. This novel approach provides a robust solution to CS ASR in unseen domains, thereby significantly enhancing our contribution to the field. By utilizing BAM to both identify and normalize CS phrases, AdaCS enhances its adaptive capabilities with a biased list of words provided during inference. Our method demonstrates impressive performance and the ability to handle unseen CS phrases across various domains. Experiments show that AdaCS outperforms previous state-of-the-art method on Vietnamese CS ASR normalization by considerable WER reduction of 56.2% and 36.8% on the two proposed test sets.
comment: Accepted at ICASSP 2025
☆ Collaborative Learning for 3D Hand-Object Reconstruction and Compositional Action Recognition from Egocentric RGB Videos Using Superquadrics AAAI 2025
With the availability of egocentric 3D hand-object interaction datasets, there is increasing interest in developing unified models for hand-object pose estimation and action recognition. However, existing methods still struggle to recognise seen actions on unseen objects due to the limitations in representing object shape and movement using 3D bounding boxes. Additionally, the reliance on object templates at test time limits their generalisability to unseen objects. To address these challenges, we propose to leverage superquadrics as an alternative 3D object representation to bounding boxes and demonstrate their effectiveness on both template-free object reconstruction and action recognition tasks. Moreover, as we find that pure appearance-based methods can outperform the unified methods, the potential benefits from 3D geometric information remain unclear. Therefore, we study the compositionality of actions by considering a more challenging task where the training combinations of verbs and nouns do not overlap with the testing split. We extend H2O and FPHA datasets with compositional splits and design a novel collaborative learning framework that can explicitly reason about the geometric relations between hands and the manipulated object. Through extensive quantitative and qualitative evaluations, we demonstrate significant improvements over the state-of-the-arts in (compositional) action recognition.
comment: Accepted to AAAI 2025
☆ MathReader : Text-to-Speech for Mathematical Documents ICASSP 2025
TTS (Text-to-Speech) document reader from Microsoft, Adobe, Apple, and OpenAI have been serviced worldwide. They provide relatively good TTS results for general plain text, but sometimes skip contents or provide unsatisfactory results for mathematical expressions. This is because most modern academic papers are written in LaTeX, and when LaTeX formulas are compiled, they are rendered as distinctive text forms within the document. However, traditional TTS document readers output only the text as it is recognized, without considering the mathematical meaning of the formulas. To address this issue, we propose MathReader, which effectively integrates OCR, a fine-tuned T5 model, and TTS. MathReader demonstrated a lower Word Error Rate (WER) than existing TTS document readers, such as Microsoft Edge and Adobe Acrobat, when processing documents containing mathematical formulas. MathReader reduced the WER from 0.510 to 0.281 compared to Microsoft Edge, and from 0.617 to 0.281 compared to Adobe Acrobat. This will significantly contribute to alleviating the inconvenience faced by users who want to listen to documents, especially those who are visually impaired. The code is available at https://github.com/hyeonsieun/MathReader.
comment: Accepted at ICASSP 2025
☆ Video Quality Assessment for Online Processing: From Spatial to Temporal Sampling
With the rapid development of multimedia processing and deep learning technologies, especially in the field of video understanding, video quality assessment (VQA) has achieved significant progress. Although researchers have moved from designing efficient video quality mapping models to various research directions, in-depth exploration of the effectiveness-efficiency trade-offs of spatio-temporal modeling in VQA models is still less sufficient. Considering the fact that videos have highly redundant information, this paper investigates this problem from the perspective of joint spatial and temporal sampling, aiming to seek the answer to how little information we should keep at least when feeding videos into the VQA models while with acceptable performance sacrifice. To this end, we drastically sample the video's information from both spatial and temporal dimensions, and the heavily squeezed video is then fed into a stable VQA model. Comprehensive experiments regarding joint spatial and temporal sampling are conducted on six public video quality databases, and the results demonstrate the acceptable performance of the VQA model when throwing away most of the video information. Furthermore, with the proposed joint spatial and temporal sampling strategy, we make an initial attempt to design an online VQA model, which is instantiated by as simple as possible a spatial feature extractor, a temporal feature fusion module, and a global quality regression module. Through quantitative and qualitative experiments, we verify the feasibility of online VQA model by simplifying itself and reducing input.
☆ ADKGD: Anomaly Detection in Knowledge Graphs with Dual-Channel Training
In the current development of large language models (LLMs), it is important to ensure the accuracy and reliability of the underlying data sources. LLMs are critical for various applications, but they often suffer from hallucinations and inaccuracies due to knowledge gaps in the training data. Knowledge graphs (KGs), as a powerful structural tool, could serve as a vital external information source to mitigate the aforementioned issues. By providing a structured and comprehensive understanding of real-world data, KGs enhance the performance and reliability of LLMs. However, it is common that errors exist in KGs while extracting triplets from unstructured data to construct KGs. This could lead to degraded performance in downstream tasks such as question-answering and recommender systems. Therefore, anomaly detection in KGs is essential to identify and correct these errors. This paper presents an anomaly detection algorithm in knowledge graphs with dual-channel learning (ADKGD). ADKGD leverages a dual-channel learning approach to enhance representation learning from both the entity-view and triplet-view perspectives. Furthermore, using a cross-layer approach, our framework integrates internal information aggregation and context information aggregation. We introduce a kullback-leibler (KL)-loss component to improve the accuracy of the scoring function between the dual channels. To evaluate ADKGD's performance, we conduct empirical studies on three real-world KGs: WN18RR, FB15K, and NELL-995. Experimental results demonstrate that ADKGD outperforms the state-of-the-art anomaly detection algorithms. The source code and datasets are publicly available at https://github.com/csjywu1/ADKGD.
comment: Preprint. 11 figures, 6 tables
☆ Representation Learning of Point Cloud Upsampling in Global and Local Inputs
In recent years, point cloud upsampling has been widely applied in fields such as 3D reconstruction. Our study investigates the factors influencing point cloud upsampling on both global and local levels through representation learning. Specifically, the paper inputs global and local information of the same point cloud model object into two encoders to extract these features, fuses them, and then feeds the combined features into an upsampling decoder. The goal is to address issues of sparsity and noise in point clouds by leveraging prior knowledge from both global and local inputs. And the proposed framework can be applied to any state-of-the-art point cloud upsampling neural network. Experiments were conducted on a series of autoencoder-based models utilizing deep learning, yielding interpretability for both global and local inputs, and it has been proven in the results that our proposed framework can further improve the upsampling effect in previous SOTA works. At the same time, the Saliency Map reflects the differences between global and local feature inputs, as well as the effectiveness of training with both inputs in parallel.
☆ Value Compass Leaderboard: A Platform for Fundamental and Validated Evaluation of LLMs Values
As Large Language Models (LLMs) achieve remarkable breakthroughs, aligning their values with humans has become imperative for their responsible development and customized applications. However, there still lack evaluations of LLMs values that fulfill three desirable goals. (1) Value Clarification: We expect to clarify the underlying values of LLMs precisely and comprehensively, while current evaluations focus narrowly on safety risks such as bias and toxicity. (2) Evaluation Validity: Existing static, open-source benchmarks are prone to data contamination and quickly become obsolete as LLMs evolve. Additionally, these discriminative evaluations uncover LLMs' knowledge about values, rather than valid assessments of LLMs' behavioral conformity to values. (3) Value Pluralism: The pluralistic nature of human values across individuals and cultures is largely ignored in measuring LLMs value alignment. To address these challenges, we presents the Value Compass Leaderboard, with three correspondingly designed modules. It (i) grounds the evaluation on motivationally distinct \textit{basic values to clarify LLMs' underlying values from a holistic view; (ii) applies a \textit{generative evolving evaluation framework with adaptive test items for evolving LLMs and direct value recognition from behaviors in realistic scenarios; (iii) propose a metric that quantifies LLMs alignment with a specific value as a weighted sum over multiple dimensions, with weights determined by pluralistic values.
☆ Logic Meets Magic: LLMs Cracking Smart Contract Vulnerabilities
Smart contract vulnerabilities caused significant economic losses in blockchain applications. Large Language Models (LLMs) provide new possibilities for addressing this time-consuming task. However, state-of-the-art LLM-based detection solutions are often plagued by high false-positive rates. In this paper, we push the boundaries of existing research in two key ways. First, our evaluation is based on Solidity v0.8, offering the most up-to-date insights compared to prior studies that focus on older versions (v0.4). Second, we leverage the latest five LLM models (across companies), ensuring comprehensive coverage across the most advanced capabilities in the field. We conducted a series of rigorous evaluations. Our experiments demonstrate that a well-designed prompt can reduce the false-positive rate by over 60%. Surprisingly, we also discovered that the recall rate for detecting some specific vulnerabilities in Solidity v0.8 has dropped to just 13% compared to earlier versions (i.e., v0.4). Further analysis reveals the root cause of this decline: the reliance of LLMs on identifying changes in newly introduced libraries and frameworks during detection.
☆ PoAct: Policy and Action Dual-Control Agent for Generalized Applications
Based on their superior comprehension and reasoning capabilities, Large Language Model (LLM) driven agent frameworks have achieved significant success in numerous complex reasoning tasks. ReAct-like agents can solve various intricate problems step-by-step through progressive planning and tool calls, iteratively optimizing new steps based on environmental feedback. However, as the planning capabilities of LLMs improve, the actions invoked by tool calls in ReAct-like frameworks often misalign with complex planning and challenging data organization. Code Action addresses these issues while also introducing the challenges of a more complex action space and more difficult action organization. To leverage Code Action and tackle the challenges of its complexity, this paper proposes Policy and Action Dual-Control Agent (PoAct) for generalized applications. The aim is to achieve higher-quality code actions and more accurate reasoning paths by dynamically switching reasoning policies and modifying the action space. Experimental results on the Agent Benchmark for both legal and generic scenarios demonstrate the superior reasoning capabilities and reduced token consumption of our approach in complex tasks. On the LegalAgentBench, our method shows a 20 percent improvement over the baseline while requiring fewer tokens. We conducted experiments and analyses on the GPT-4o and GLM-4 series models, demonstrating the significant potential and scalability of our approach to solve complex problems.
☆ Unveiling the Potential of Text in High-Dimensional Time Series Forecasting NeurIPS24
Time series forecasting has traditionally focused on univariate and multivariate numerical data, often overlooking the benefits of incorporating multimodal information, particularly textual data. In this paper, we propose a novel framework that integrates time series models with Large Language Models to improve high-dimensional time series forecasting. Inspired by multimodal models, our method combines time series and textual data in the dual-tower structure. This fusion of information creates a comprehensive representation, which is then processed through a linear layer to generate the final forecast. Extensive experiments demonstrate that incorporating text enhances high-dimensional time series forecasting performance. This work paves the way for further research in multimodal time series forecasting.
comment: Accepted by NeurIPS24 TSALM Workshop
☆ ACCon: Angle-Compensated Contrastive Regularizer for Deep Regression AAAI-2025
In deep regression, capturing the relationship among continuous labels in feature space is a fundamental challenge that has attracted increasing interest. Addressing this issue can prevent models from converging to suboptimal solutions across various regression tasks, leading to improved performance, especially for imbalanced regression and under limited sample sizes. However, existing approaches often rely on order-aware representation learning or distance-based weighting. In this paper, we hypothesize a linear negative correlation between label distances and representation similarities in regression tasks. To implement this, we propose an angle-compensated contrastive regularizer for deep regression, which adjusts the cosine distance between anchor and negative samples within the contrastive learning framework. Our method offers a plug-and-play compatible solution that extends most existing contrastive learning methods for regression tasks. Extensive experiments and theoretical analysis demonstrate that our proposed angle-compensated contrastive regularizer not only achieves competitive regression performance but also excels in data efficiency and effectiveness on imbalanced datasets.
comment: Accept by AAAI-2025 (The 39th Annual AAAI Conference on Artificial Intelligence)
☆ A Proposed Large Language Model-Based Smart Search for Archive System
This study presents a novel framework for smart search in digital archival systems, leveraging the capabilities of Large Language Models (LLMs) to enhance information retrieval. By employing a Retrieval-Augmented Generation (RAG) approach, the framework enables the processing of natural language queries and transforming non-textual data into meaningful textual representations. The system integrates advanced metadata generation techniques, a hybrid retrieval mechanism, a router query engine, and robust response synthesis, the results proved search precision and relevance. We present the architecture and implementation of the system and evaluate its performance in four experiments concerning LLM efficiency, hybrid retrieval optimizations, multilingual query handling, and the impacts of individual components. Obtained results show significant improvements over conventional approaches and have demonstrated the potential of AI-powered systems to transform modern archival practices.
comment: The 13th International Symposium on Information and Communication Technology (SOICT 2024)
☆ Neural Probabilistic Circuits: Enabling Compositional and Interpretable Predictions through Logical Reasoning
End-to-end deep neural networks have achieved remarkable success across various domains but are often criticized for their lack of interpretability. While post hoc explanation methods attempt to address this issue, they often fail to accurately represent these black-box models, resulting in misleading or incomplete explanations. To overcome these challenges, we propose an inherently transparent model architecture called Neural Probabilistic Circuits (NPCs), which enable compositional and interpretable predictions through logical reasoning. In particular, an NPC consists of two modules: an attribute recognition model, which predicts probabilities for various attributes, and a task predictor built on a probabilistic circuit, which enables logical reasoning over recognized attributes to make class predictions. To train NPCs, we introduce a three-stage training algorithm comprising attribute recognition, circuit construction, and joint optimization. Moreover, we theoretically demonstrate that an NPC's error is upper-bounded by a linear combination of the errors from its modules. To further demonstrate the interpretability of NPC, we provide both the most probable explanations and the counterfactual explanations. Empirical results on four benchmark datasets show that NPCs strike a balance between interpretability and performance, achieving results competitive even with those of end-to-end black-box models while providing enhanced interpretability.
☆ ViSoLex: An Open-Source Repository for Vietnamese Social Media Lexical Normalization COLING 2025
ViSoLex is an open-source system designed to address the unique challenges of lexical normalization for Vietnamese social media text. The platform provides two core services: Non-Standard Word (NSW) Lookup and Lexical Normalization, enabling users to retrieve standard forms of informal language and standardize text containing NSWs. ViSoLex's architecture integrates pre-trained language models and weakly supervised learning techniques to ensure accurate and efficient normalization, overcoming the scarcity of labeled data in Vietnamese. This paper details the system's design, functionality, and its applications for researchers and non-technical users. Additionally, ViSoLex offers a flexible, customizable framework that can be adapted to various datasets and research requirements. By publishing the source code, ViSoLex aims to contribute to the development of more robust Vietnamese natural language processing tools and encourage further research in lexical normalization. Future directions include expanding the system's capabilities for additional languages and improving the handling of more complex non-standard linguistic patterns.
comment: The 31st International Conference on Computational Linguistics (COLING 2025)
☆ UNetVL: Enhancing 3D Medical Image Segmentation with Chebyshev KAN Powered Vision-LSTM
3D medical image segmentation has progressed considerably due to Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), yet these methods struggle to balance long-range dependency acquisition with computational efficiency. To address this challenge, we propose UNETVL (U-Net Vision-LSTM), a novel architecture that leverages recent advancements in temporal information processing. UNETVL incorporates Vision-LSTM (ViL) for improved scalability and memory functions, alongside an efficient Chebyshev Kolmogorov-Arnold Networks (KAN) to handle complex and long-range dependency patterns more effectively. We validated our method on the ACDC and AMOS2022 (post challenge Task 2) benchmark datasets, showing a significant improvement in mean Dice score compared to recent state-of-the-art approaches, especially over its predecessor, UNETR, with increases of 7.3% on ACDC and 15.6% on AMOS, respectively. Extensive ablation studies were conducted to demonstrate the impact of each component in UNETVL, providing a comprehensive understanding of its architecture. Our code is available at https://github.com/tgrex6/UNETVL, facilitating further research and applications in this domain.
☆ A Multi-Modal Deep Learning Framework for Pan-Cancer Prognosis
Prognostic task is of great importance as it closely related to the survival analysis of patients, the optimization of treatment plans and the allocation of resources. The existing prognostic models have shown promising results on specific datasets, but there are limitations in two aspects. On the one hand, they merely explore certain types of modal data, such as patient histopathology WSI and gene expression analysis. On the other hand, they adopt the per-cancer-per-model paradigm, which means the trained models can only predict the prognostic effect of a single type of cancer, resulting in weak generalization ability. In this paper, a deep-learning based model, named UMPSNet, is proposed. Specifically, to comprehensively understand the condition of patients, in addition to constructing encoders for histopathology images and genomic expression profiles respectively, UMPSNet further integrates four types of important meta data (demographic information, cancer type information, treatment protocols, and diagnosis results) into text templates, and then introduces a text encoder to extract textual features. In addition, the optimal transport OT-based attention mechanism is utilized to align and fuse features of different modalities. Furthermore, a guided soft mixture of experts (GMoE) mechanism is introduced to effectively address the issue of distribution differences among multiple cancer datasets. By incorporating the multi-modality of patient data and joint training, UMPSNet outperforms all SOTA approaches, and moreover, it demonstrates the effectiveness and generalization ability of the proposed learning paradigm of a single model for multiple cancer types. The code of UMPSNet is available at https://github.com/binging512/UMPSNet.
☆ AlgoRxplorers | Precision in Mutation -- Enhancing Drug Design with Advanced Protein Stability Prediction Tools
Predicting the impact of single-point amino acid mutations on protein stability is essential for understanding disease mechanisms and advancing drug development. Protein stability, quantified by changes in Gibbs free energy ($\Delta\Delta G$), is influenced by these mutations. However, the scarcity of data and the complexity of model interpretation pose challenges in accurately predicting stability changes. This study proposes the application of deep neural networks, leveraging transfer learning and fusing complementary information from different models, to create a feature-rich representation of the protein stability landscape. We developed four models, with our third model, ThermoMPNN+, demonstrating the best performance in predicting $\Delta\Delta G$ values. This approach, which integrates diverse feature sets and embeddings through latent transfusion techniques, aims to refine $\Delta\Delta G$ predictions and contribute to a deeper understanding of protein dynamics, potentially leading to advancements in disease research and drug discovery.
☆ Likelihood Training of Cascaded Diffusion Models via Hierarchical Volume-preserving Maps ICLR 2024
Cascaded models are multi-scale generative models with a marked capacity for producing perceptually impressive samples at high resolutions. In this work, we show that they can also be excellent likelihood models, so long as we overcome a fundamental difficulty with probabilistic multi-scale models: the intractability of the likelihood function. Chiefly, in cascaded models each intermediary scale introduces extraneous variables that cannot be tractably marginalized out for likelihood evaluation. This issue vanishes by modeling the diffusion process on latent spaces induced by a class of transformations we call hierarchical volume-preserving maps, which decompose spatially structured data in a hierarchical fashion without introducing local distortions in the latent space. We demonstrate that two such maps are well-known in the literature for multiscale modeling: Laplacian pyramids and wavelet transforms. Not only do such reparameterizations allow the likelihood function to be directly expressed as a joint likelihood over the scales, we show that the Laplacian pyramid and wavelet transform also produces significant improvements to the state-of-the-art on a selection of benchmarks in likelihood modeling, including density estimation, lossless compression, and out-of-distribution detection. Investigating the theoretical basis of our empirical gains we uncover deep connections to score matching under the Earth Mover's Distance (EMD), which is a well-known surrogate for perceptual similarity. Code can be found at \href{https://github.com/lihenryhfl/pcdm}{this https url}.
comment: Spotlight at ICLR 2024
☆ Motion Tracks: A Unified Representation for Human-Robot Transfer in Few-Shot Imitation Learning
Teaching robots to autonomously complete everyday tasks remains a challenge. Imitation Learning (IL) is a powerful approach that imbues robots with skills via demonstrations, but is limited by the labor-intensive process of collecting teleoperated robot data. Human videos offer a scalable alternative, but it remains difficult to directly train IL policies from them due to the lack of robot action labels. To address this, we propose to represent actions as short-horizon 2D trajectories on an image. These actions, or motion tracks, capture the predicted direction of motion for either human hands or robot end-effectors. We instantiate an IL policy called Motion Track Policy (MT-pi) which receives image observations and outputs motion tracks as actions. By leveraging this unified, cross-embodiment action space, MT-pi completes tasks with high success given just minutes of human video and limited additional robot demonstrations. At test time, we predict motion tracks from two camera views, recovering 6DoF trajectories via multi-view synthesis. MT-pi achieves an average success rate of 86.5% across 4 real-world tasks, outperforming state-of-the-art IL baselines which do not leverage human data or our action space by 40%, and generalizes to scenarios seen only in human videos. Code and videos are available on our website https://portal-cornell.github.io/motion_track_policy/.
☆ Graph Contrastive Learning on Multi-label Classification for Recommendations
In business analysis, providing effective recommendations is essential for enhancing company profits. The utilization of graph-based structures, such as bipartite graphs, has gained popularity for their ability to analyze complex data relationships. Link prediction is crucial for recommending specific items to users. Traditional methods in this area often involve identifying patterns in the graph structure or using representational techniques like graph neural networks (GNNs). However, these approaches encounter difficulties as the volume of data increases. To address these challenges, we propose a model called Graph Contrastive Learning for Multi-label Classification (MCGCL). MCGCL leverages contrastive learning to enhance recommendation effectiveness. The model incorporates two training stages: a main task and a subtask. The main task is holistic user-item graph learning to capture user-item relationships. The homogeneous user-user (item-item) subgraph is constructed to capture user-user and item-item relationships in the subtask. We assessed the performance using real-world datasets from Amazon Reviews in multi-label classification tasks. Comparative experiments with state-of-the-art methods confirm the effectiveness of MCGCL, highlighting its potential for improving recommendation systems.
comment: Preprint. 10 figures, 5 tables
☆ Data Enrichment Work and AI Labor in Latin America and the Caribbean
The global AI surge demands crowdworkers from diverse languages and cultures. They are pivotal in labeling data for enabling global AI systems. Despite global significance, research has primarily focused on understanding the perspectives and experiences of US and India crowdworkers, leaving a notable gap. To bridge this, we conducted a survey with 100 crowdworkers across 16 Latin American and Caribbean countries. We discovered that these workers exhibited pride and respect for their digital labor, with strong support and admiration from their families. Notably, crowd work was also seen as a stepping stone to financial and professional independence. Surprisingly, despite wanting more connection, these workers also felt isolated from peers and doubtful of others' labor quality. They resisted collaboration and gender-based tools, valuing gender-neutrality. Our work advances HCI understanding of Latin American and Caribbean crowdwork, offering insights for digital resistance tools for the region.
comment: 17 pages of content with 2 figures
☆ Combining LLM decision and RL action selection to improve RL policy for adaptive interventions
Reinforcement learning (RL) is increasingly being used in the healthcare domain, particularly for the development of personalized health adaptive interventions. Inspired by the success of Large Language Models (LLMs), we are interested in using LLMs to update the RL policy in real time, with the goal of accelerating personalization. We use the text-based user preference to influence the action selection on the fly, in order to immediately incorporate the user preference. We use the term "user preference" as a broad term to refer to a user personal preference, constraint, health status, or a statement expressing like or dislike, etc. Our novel approach is a hybrid method that combines the LLM response and the RL action selection to improve the RL policy. Given an LLM prompt that incorporates the user preference, the LLM acts as a filter in the typical RL action selection. We investigate different prompting strategies and action selection strategies. To evaluate our approach, we implement a simulation environment that generates the text-based user preferences and models the constraints that impact behavioral dynamics. We show that our approach is able to take into account the text-based user preferences, while improving the RL policy, thus improving personalization in adaptive intervention.
♻ ☆ Few-Shot Task Learning through Inverse Generative Modeling
Learning the intents of an agent, defined by its goals or motion style, is often extremely challenging from just a few examples. We refer to this problem as task concept learning and present our approach, Few-Shot Task Learning through Inverse Generative Modeling (FTL-IGM), which learns new task concepts by leveraging invertible neural generative models. The core idea is to pretrain a generative model on a set of basic concepts and their demonstrations. Then, given a few demonstrations of a new concept (such as a new goal or a new action), our method learns the underlying concepts through backpropagation without updating the model weights, thanks to the invertibility of the generative model. We evaluate our method in five domains -- object rearrangement, goal-oriented navigation, motion caption of human actions, autonomous driving, and real-world table-top manipulation. Our experimental results demonstrate that via the pretrained generative model, we successfully learn novel concepts and generate agent plans or motion corresponding to these concepts in (1) unseen environments and (2) in composition with training concepts.
comment: Added acknowledgment
♻ ☆ Harnessing Multimodal Large Language Models for Multimodal Sequential Recommendation
Recent advances in Large Language Models (LLMs) have demonstrated significant potential in the field of Recommendation Systems (RSs). Most existing studies have focused on converting user behavior logs into textual prompts and leveraging techniques such as prompt tuning to enable LLMs for recommendation tasks. Meanwhile, research interest has recently grown in multimodal recommendation systems that integrate data from images, text, and other sources using modality fusion techniques. This introduces new challenges to the existing LLM-based recommendation paradigm which relies solely on text modality information. Moreover, although Multimodal Large Language Models (MLLMs) capable of processing multi-modal inputs have emerged, how to equip MLLMs with multi-modal recommendation capabilities remains largely unexplored. To this end, in this paper, we propose the Multimodal Large Language Model-enhanced Multimodaln Sequential Recommendation (MLLM-MSR) model. To capture the dynamic user preference, we design a two-stage user preference summarization method. Specifically, we first utilize an MLLM-based item-summarizer to extract image feature given an item and convert the image into text. Then, we employ a recurrent user preference summarization generation paradigm to capture the dynamic changes in user preferences based on an LLM-based user-summarizer. Finally, to enable the MLLM for multi-modal recommendation task, we propose to fine-tune a MLLM-based recommender using Supervised Fine-Tuning (SFT) techniques. Extensive evaluations across various datasets validate the effectiveness of MLLM-MSR, showcasing its superior ability to capture and adapt to the evolving dynamics of user preferences.
♻ ☆ The importance of visual modelling languages in generative software engineering
Multimodal GPTs represent a watershed in the interplay between Software Engineering and Generative Artificial Intelligence. GPT-4 accepts image and text inputs, rather than simply natural language. We investigate relevant use cases stemming from these enhanced capabilities of GPT-4. To the best of our knowledge, no other work has investigated similar use cases involving Software Engineering tasks carried out via multimodal GPTs prompted with a mix of diagrams and natural language.
comment: 9 pages, working paper
♻ ☆ FlashRNN: Optimizing Traditional RNNs on Modern Hardware
While Transformers and other sequence-parallelizable neural network architectures seem like the current state of the art in sequence modeling, they specifically lack state-tracking capabilities. These are important for time-series tasks and logical reasoning. Traditional RNNs like LSTMs and GRUs, as well as modern variants like sLSTM do have these capabilities at the cost of strictly sequential processing. While this is often seen as a strong limitation, we show how fast these networks can get with our hardware-optimization FlashRNN in Triton and CUDA, optimizing kernels to the register level on modern GPUs. We extend traditional RNNs with a parallelization variant that processes multiple RNNs of smaller hidden state in parallel, similar to the head-wise processing in Transformers. To enable flexibility on different GPU variants, we introduce a new optimization framework for hardware-internal cache sizes, memory and compute handling. It models the hardware in a setting using polyhedral-like constraints, including the notion of divisibility. This speeds up the solution process in our ConstrINT library for general integer constraint satisfaction problems (integer CSPs). We show that our kernels can achieve 50x speed-ups over a vanilla PyTorch implementation and allow 40x larger hidden sizes compared to our Triton implementation. Our open-source kernels and the optimization library are released here to boost research in the direction of state-tracking enabled RNNs and sequence modeling: \url{https://github.com/NX-AI/flashrnn}
♻ ☆ Explainable AI for Classifying UTI Risk Groups Using a Real-World Linked EHR and Pathology Lab Dataset
The use of machine learning and AI on electronic health records (EHRs) holds substantial potential for clinical insight. However, this approach faces challenges due to data heterogeneity, sparsity, temporal misalignment, and limited labeled outcomes. In this context, we leverage a linked EHR dataset of approximately one million de-identified individuals from Bristol, North Somerset, and South Gloucestershire, UK, to characterize urinary tract infections (UTIs). We implemented a data pre-processing and curation pipeline that transforms the raw EHR data into a structured format suitable for developing predictive models focused on data fairness, accountability and transparency. Given the limited availability and biases of ground truth UTI outcomes, we introduce a UTI risk estimation framework informed by clinical expertise to estimate UTI risk across individual patient timelines. Pairwise XGBoost models are trained using this framework to differentiate UTI risk categories with explainable AI techniques applied to identify key predictors and support interpretability. Our findings reveal differences in clinical and demographic predictors across risk groups. While this study highlights the potential of AI-driven insights to support UTI clinical decision-making, further investigation of patient sub-strata and extensive validation are needed to ensure robustness and applicability in clinical practice.
♻ ☆ Small Language Models can Outperform Humans in Short Creative Writing: A Study Comparing SLMs with Humans and LLMs COLING 2025
In this paper, we evaluate the creative fiction writing abilities of a fine-tuned small language model (SLM), BART-large, and compare its performance to human writers and two large language models (LLMs): GPT-3.5 and GPT-4o. Our evaluation consists of two experiments: (i) a human study in which 68 participants rated short stories from humans and the SLM on grammaticality, relevance, creativity, and attractiveness, and (ii) a qualitative linguistic analysis examining the textual characteristics of stories produced by each model. In the first experiment, BART-large outscored average human writers overall (2.11 vs. 1.85), a 14% relative improvement, though the slight human advantage in creativity was not statistically significant. In the second experiment, qualitative analysis showed that while GPT-4o demonstrated near-perfect coherence and used less cliche phrases, it tended to produce more predictable language, with only 3% of its synopses featuring surprising associations (compared to 15% for BART). These findings highlight how model size and fine-tuning influence the balance between creativity, fluency, and coherence in creative writing tasks, and demonstrate that smaller models can, in certain contexts, rival both humans and larger models.
comment: Accepted as Main Conference Paper at COLING 2025
♻ ☆ Zero-Shot Pupil Segmentation with SAM 2: A Case Study of Over 14 Million Images
We explore the transformative potential of SAM 2, a vision foundation model, in advancing gaze estimation and eye tracking technologies. By significantly reducing annotation time, lowering technical barriers through its ease of deployment, and enhancing segmentation accuracy, SAM 2 addresses critical challenges faced by researchers and practitioners. Utilizing its zero-shot segmentation capabilities with minimal user input-a single click per video-we tested SAM 2 on over 14 million eye images from diverse datasets, including virtual reality setups and the world's largest unified dataset recorded using wearable eye trackers. Remarkably, in pupil segmentation tasks, SAM 2 matches the performance of domain-specific models trained solely on eye images, achieving competitive mean Intersection over Union (mIoU) scores of up to 93% without fine-tuning. Additionally, we provide our code and segmentation masks for these widely used datasets to promote further research.
comment: Virmarie Maquiling and Sean Anthony Byrne contributed equally to this paper, 8 pages, 3 figures, ETRA 2025, pre-print
♻ ☆ Distributed Representations Enable Robust Multi-Timescale Symbolic Computation in Neuromorphic Hardware
Programming recurrent spiking neural networks (RSNNs) to robustly perform multi-timescale computation remains a difficult challenge. To address this, we describe a single-shot weight learning scheme to embed robust multi-timescale dynamics into attractor-based RSNNs, by exploiting the properties of high-dimensional distributed representations. We embed finite state machines into the RSNN dynamics by superimposing a symmetric autoassociative weight matrix and asymmetric transition terms, which are each formed by the vector binding of an input and heteroassociative outer-products between states. Our approach is validated through simulations with highly nonideal weights; an experimental closed-loop memristive hardware setup; and on Loihi 2, where it scales seamlessly to large state machines. This work introduces a scalable approach to embed robust symbolic computation through recurrent dynamics into neuromorphic hardware, without requiring parameter fine-tuning or significant platform-specific optimisation. Moreover, it demonstrates that distributed symbolic representations serve as a highly capable representation-invariant language for cognitive algorithms in neuromorphic hardware.
comment: 19 pages, 7 figures. Supplementary material: 13 pages, 8 figures. Accepted for publication in Neuromorphic Computing and Engineering
♻ ☆ Constructing and explaining machine learning models for chemistry: example of the exploration and design of boron-based Lewis acids
The integration of machine learning (ML) into chemistry offers transformative potential in the design of molecules with targeted properties. However, the focus has often been on creating highly efficient predictive models, sometimes at the expense of interpretability. In this study, we leverage explainable AI techniques to explore the rational design of boron-based Lewis acids, which play a pivotal role in organic reactions due to their electron-ccepting properties. Using Fluoride Ion Affinity as a proxy for Lewis acidity, we developed interpretable ML models based on chemically meaningful descriptors, including ab initio computed features and substituent-based parameters derived from the Hammett linear free-energy relationship. By constraining the chemical space to well-defined molecular scaffolds, we achieved highly accurate predictions (mean absolute error < 6 kJ/mol), surpassing conventional black-box deep learning models in low-data regimes. Interpretability analyses of the models shed light on the origin of Lewis acidity in these compounds and identified actionable levers to modulate it through the nature and positioning of substituents on the molecular scaffold. This work bridges ML and chemist's way of thinking, demonstrating how explainable models can inspire molecular design and enhance scientific understanding of chemical reactivity.
comment: Main text is 14 pages, 7 figures, 1 scheme. Supporting information is 25 pages. For associated code and datasets, see https://github.com/jfenogli/XAI_boron_LA
♻ ☆ Project Tracyn: Generative Artificial Intelligence based Peripherals Trace Synthesizer
Peripheral Component Interconnect Express (PCIe) is the de facto interconnect standard for high-speed peripherals and CPUs. Prototyping and optimizing PCIe devices for emerging scenarios is an ongoing challenge. Since Transaction Layer Packets (TLPs) capture device-CPU interactions, it is crucial to analyze and generate realistic TLP traces for effective device design and optimization. Generative AI offers a promising approach for creating intricate, custom TLP traces necessary for PCIe hardware and software development. However, existing models often generate impractical traces due to the absence of PCIe-specific constraints, such as TLP ordering and causality. This paper presents Phantom, the first framework that treats TLP trace generation as a generative AI problem while incorporating PCIe-specific constraints. We validate Phantom's effectiveness by generating TLP traces for an actual PCIe network interface card. Experimental results show that Phantom produces practical, large-scale TLP traces, significantly outperforming existing models, with improvements of up to 1000$\times$ in task-specific metrics and up to 2.19$\times$ in Frechet Inception Distance (FID) compared to backbone-only methods.
♻ ☆ Mitigating Out-of-Entity Errors in Named Entity Recognition: A Sentence-Level Strategy COLING 2025
Many previous models of named entity recognition (NER) suffer from the problem of Out-of-Entity (OOE), i.e., the tokens in the entity mentions of the test samples have not appeared in the training samples, which hinders the achievement of satisfactory performance. To improve OOE-NER performance, in this paper, we propose a new framework, namely S+NER, which fully leverages sentence-level information. Our S+NER achieves better OOE-NER performance mainly due to the following two particular designs. 1) It first exploits the pre-trained language model's capability of understanding the target entity's sentence-level context with a template set. 2) Then, it refines the sentence-level representation based on the positive and negative templates, through a contrastive learning strategy and template pooling method, to obtain better NER results. Our extensive experiments on five benchmark datasets have demonstrated that, our S+NER outperforms some state-of-the-art OOE-NER models.
comment: Accepted by COLING 2025
♻ ☆ QuadWBG: Generalizable Quadrupedal Whole-Body Grasping
Legged robots with advanced manipulation capabilities have the potential to significantly improve household duties and urban maintenance. Despite considerable progress in developing robust locomotion and precise manipulation methods, seamlessly integrating these into cohesive whole-body control for real-world applications remains challenging. In this paper, we present a modular framework for robust and generalizable whole-body loco-manipulation controller based on a single arm-mounted camera. By using reinforcement learning (RL), we enable a robust low-level policy for command execution over 5 dimensions (5D) and a grasp-aware high-level policy guided by a novel metric, Generalized Oriented Reachability Map (GORM). The proposed system achieves state-of-the-art one-time grasping accuracy of 89% in the real world, including challenging tasks such as grasping transparent objects. Through extensive simulations and real-world experiments, we demonstrate that our system can effectively manage a large workspace, from floor level to above body height, and perform diverse whole-body loco-manipulation tasks.
♻ ☆ SCC-YOLO: An Improved Object Detector for Assisting in Brain Tumor Diagnosis
Brain tumors can result in neurological dysfunction, alterations in cognitive and psychological states, increased intracranial pressure, and the occurrence of seizures, thereby presenting a substantial risk to human life and health. The You Only Look Once(YOLO) series models have demonstrated superior accuracy in object detection for medical imaging. In this paper, we develop a novel SCC-YOLO architecture by integrating the SCConv attention mechanism into YOLOv9. The SCConv module reconstructs an efficient convolutional module by reducing spatial and channel redundancy among features, thereby enhancing the learning of image features. We investigate the impact of intergrating different attention mechanisms with the YOLOv9 model on brain tumor image detection using both the Br35H dataset and our self-made dataset(Brain_Tumor_Dataset). Experimental results show that on the Br35H dataset, SCC-YOLO achieved a 0.3% improvement in mAp50 compared to YOLOv9, while on our self-made dataset, SCC-YOLO exhibited a 0.5% improvement over YOLOv9. SCC-YOLO has reached state-of-the-art performance in brain tumor detection. Source code is available at : https://jihulab.com/healthcare-information-studio/SCC-YOLO/-/tree/master
♻ ☆ AI-Driven Early Mental Health Screening: Analyzing Selfies of Pregnant Women ALT
Major Depressive Disorder and anxiety disorders affect millions globally, contributing significantly to the burden of mental health issues. Early screening is crucial for effective intervention, as timely identification of mental health issues can significantly improve treatment outcomes. Artificial intelligence (AI) can be valuable for improving the screening of mental disorders, enabling early intervention and better treatment outcomes. AI-driven screening can leverage the analysis of multiple data sources, including facial features in digital images. However, existing methods often rely on controlled environments or specialized equipment, limiting their broad applicability. This study explores the potential of AI models for ubiquitous depression-anxiety screening given face-centric selfies. The investigation focuses on high-risk pregnant patients, a population that is particularly vulnerable to mental health issues. To cope with limited training data resulting from our clinical setup, pre-trained models were utilized in two different approaches: fine-tuning convolutional neural networks (CNNs) originally designed for facial expression recognition and employing vision-language models (VLMs) for zero-shot analysis of facial expressions. Experimental results indicate that the proposed VLM-based method significantly outperforms CNNs, achieving an accuracy of 77.6%. Although there is significant room for improvement, the results suggest that VLMs can be a promising approach for mental health screening.
comment: This article has been accepted for publication in HEALTHINF25 at the 18th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2025)
♻ ☆ DrLLM: Prompt-Enhanced Distributed Denial-of-Service Resistance Method with Large Language Models ICASSP2025
The increasing number of Distributed Denial of Service (DDoS) attacks poses a major threat to the Internet, highlighting the importance of DDoS mitigation. Most existing approaches require complex training methods to learn data features, which increases the complexity and generality of the application. In this paper, we propose DrLLM, which aims to mine anomalous traffic information in zero-shot scenarios through Large Language Models (LLMs). To bridge the gap between DrLLM and existing approaches, we embed the global and local information of the traffic data into the reasoning paradigm and design three modules, namely Knowledge Embedding, Token Embedding, and Progressive Role Reasoning, for data representation and reasoning. In addition we explore the generalization of prompt engineering in the cybersecurity domain to improve the classification capability of DrLLM. Our ablation experiments demonstrate the applicability of DrLLM in zero-shot scenarios and further demonstrate the potential of LLMs in the network domains. DrLLM implementation code has been open-sourced at https://github.com/liuup/DrLLM.
comment: Accepted by ICASSP2025
♻ ☆ Multi-Head Explainer: A General Framework to Improve Explainability in CNNs and Transformers
In this study, we introduce the Multi-Head Explainer (MHEX), a versatile and modular framework that enhances both the explainability and accuracy of Convolutional Neural Networks (CNNs) and Transformer-based models. MHEX consists of three core components: an Attention Gate that dynamically highlights task-relevant features, Deep Supervision that guides early layers to capture fine-grained details pertinent to the target class, and an Equivalent Matrix that unifies refined local and global representations to generate comprehensive saliency maps. Our approach demonstrates superior compatibility, enabling effortless integration into existing residual networks like ResNet and Transformer architectures such as BERT with minimal modifications. Extensive experiments on benchmark datasets in medical imaging and text classification show that MHEX not only improves classification accuracy but also produces highly interpretable and detailed saliency scores.
♻ ☆ Tiny Models are the Computational Saver for Large Models
This paper introduces TinySaver, an early-exit-like dynamic model compression approach which employs tiny models to substitute large models adaptively. Distinct from traditional compression techniques, dynamic methods like TinySaver can leverage the difficulty differences to allow certain inputs to complete their inference processes early, thereby conserving computational resources. Most existing early exit designs are implemented by attaching additional network branches to the model's backbone. Our study, however, reveals that completely independent tiny models can replace a substantial portion of the larger models' job with minimal impact on performance. Employing them as the first exit can remarkably enhance computational efficiency. By searching and employing the most appropriate tiny model as the computational saver for a given large model, the proposed approaches work as a novel and generic method to model compression. This finding will help the research community in exploring new compression methods to address the escalating computational demands posed by rapidly evolving AI models. Our evaluation of this approach in ImageNet-1k classification demonstrates its potential to reduce the number of compute operations by up to 90\%, with only negligible losses in performance, across various modern vision models.
♻ ☆ Imitating from auxiliary imperfect demonstrations via Adversarial Density Weighted Regression
We propose a novel one-step supervised imitation learning (IL) framework called Adversarial Density Regression (ADR). This IL framework aims to correct the policy learned on unknown-quality to match the expert distribution by utilizing demonstrations, without relying on the Bellman operator. Specifically, ADR addresses several limitations in previous IL algorithms: First, most IL algorithms are based on the Bellman operator, which inevitably suffer from cumulative offsets from sub-optimal rewards during multi-step update processes. Additionally, off-policy training frameworks suffer from Out-of-Distribution (OOD) state-actions. Second, while conservative terms help solve the OOD issue, balancing the conservative term is difficult. To address these limitations, we fully integrate a one-step density-weighted Behavioral Cloning (BC) objective for IL with auxiliary imperfect demonstration. Theoretically, we demonstrate that this adaptation can effectively correct the distribution of policies trained on unknown-quality datasets to align with the expert policy's distribution. Moreover, the difference between the empirical and the optimal value function is proportional to the upper bound of ADR's objective, indicating that minimizing ADR's objective is akin to approaching the optimal value. Experimentally, we validated the performance of ADR by conducting extensive evaluations. Specifically, ADR outperforms all of the selected IL algorithms on tasks from the Gym-Mujoco domain. Meanwhile, it achieves an 89.5% improvement over IQL when utilizing ground truth rewards on tasks from the Adroit and Kitchen domains. Our codebase will be released at: https://github.com/stevezhangzA/Adverserial_Density_Regression.
♻ ☆ D3RM: A Discrete Denoising Diffusion Refinement Model for Piano Transcription ICASSP 2025
Diffusion models have been widely used in the generative domain due to their convincing performance in modeling complex data distributions. Moreover, they have shown competitive results on discriminative tasks, such as image segmentation. While diffusion models have also been explored for automatic music transcription, their performance has yet to reach a competitive level. In this paper, we focus on discrete diffusion model's refinement capabilities and present a novel architecture for piano transcription. Our model utilizes Neighborhood Attention layers as the denoising module, gradually predicting the target high-resolution piano roll, conditioned on the finetuned features of a pretrained acoustic model. To further enhance refinement, we devise a novel strategy which applies distinct transition states during training and inference stage of discrete diffusion models. Experiments on the MAESTRO dataset show that our approach outperforms previous diffusion-based piano transcription models and the baseline model in terms of F1 score. Our code is available in https://github.com/hanshounsu/d3rm.
comment: Accepted to ICASSP 2025
♻ ☆ Are LLMs Good Cryptic Crossword Solvers?
Cryptic crosswords are puzzles that rely not only on general knowledge but also on the solver's ability to manipulate language on different levels and deal with various types of wordplay. Previous research suggests that solving such puzzles is a challenge even for modern NLP models. However, the abilities of large language models (LLMs) have not yet been tested on this task. In this paper, we establish the benchmark results for three popular LLMs -- LLaMA2, Mistral, and ChatGPT -- showing that their performance on this task is still far from that of humans.
♻ ☆ SyncDiff: Synchronized Motion Diffusion for Multi-Body Human-Object Interaction Synthesis
Synthesizing realistic human-object interaction motions is a critical problem in VR/AR and human animation. Unlike the commonly studied scenarios involving a single human or hand interacting with one object, we address a more generic multi-body setting with arbitrary numbers of humans, hands, and objects. This complexity introduces significant challenges in synchronizing motions due to the high correlations and mutual influences among bodies. To address these challenges, we introduce SyncDiff, a novel method for multi-body interaction synthesis using a synchronized motion diffusion strategy. SyncDiff employs a single diffusion model to capture the joint distribution of multi-body motions. To enhance motion fidelity, we propose a frequency-domain motion decomposition scheme. Additionally, we introduce a new set of alignment scores to emphasize the synchronization of different body motions. SyncDiff jointly optimizes both data sample likelihood and alignment likelihood through an explicit synchronization strategy. Extensive experiments across four datasets with various multi-body configurations demonstrate the superiority of SyncDiff over existing state-of-the-art motion synthesis methods.
♻ ☆ VaeDiff-DocRE: End-to-end Data Augmentation Framework for Document-level Relation Extraction COLING 2025
Document-level Relation Extraction (DocRE) aims to identify relationships between entity pairs within a document. However, most existing methods assume a uniform label distribution, resulting in suboptimal performance on real-world, imbalanced datasets. To tackle this challenge, we propose a novel data augmentation approach using generative models to enhance data from the embedding space. Our method leverages the Variational Autoencoder (VAE) architecture to capture all relation-wise distributions formed by entity pair representations and augment data for underrepresented relations. To better capture the multi-label nature of DocRE, we parameterize the VAE's latent space with a Diffusion Model. Additionally, we introduce a hierarchical training framework to integrate the proposed VAE-based augmentation module into DocRE systems. Experiments on two benchmark datasets demonstrate that our method outperforms state-of-the-art models, effectively addressing the long-tail distribution problem in DocRE.
comment: COLING 2025
♻ ☆ PSA-VLM: Enhancing Vision-Language Model Safety through Progressive Concept-Bottleneck-Driven Alignment
Benefiting from the powerful capabilities of Large Language Models (LLMs), pre-trained visual encoder models connected to LLMs form Vision Language Models (VLMs). However, recent research shows that the visual modality in VLMs is highly vulnerable, allowing attackers to bypass safety alignment in LLMs through visually transmitted content, launching harmful attacks. To address this challenge, we propose a progressive concept-based alignment strategy, PSA-VLM, which incorporates safety modules as concept bottlenecks to enhance visual modality safety alignment. By aligning model predictions with specific safety concepts, we improve defenses against risky images, enhancing explainability and controllability while minimally impacting general performance. Our method is obtained through two-stage training. The low computational cost of the first stage brings very effective performance improvement, and the fine-tuning of the language model in the second stage further improves the safety performance. Our method achieves state-of-the-art results on popular VLM safety benchmark.
comment: arXiv admin note: substantial text overlap with arXiv:2405.13581
♻ ☆ Migician: Revealing the Magic of Free-Form Multi-Image Grounding in Multimodal Large Language Models
The recent advancement of Multimodal Large Language Models (MLLMs) has significantly improved their fine-grained perception of single images and general comprehension across multiple images. However, existing MLLMs still face challenges in achieving precise grounding in complex multi-image scenarios. To address this, we first explore a Chain-of-Thought (CoT) framework that integrates single-image grounding with multi-image comprehension. While partially effective, it remains unstable and struggles to capture abstract visual information due to its non-end-to-end nature. Therefore, we introduce Migician, the first multi-image grounding model capable of performing free-form and accurate grounding across multiple images. To support this, we present the MGrounding-630k dataset, which comprises data for several multi-image grounding tasks derived from existing datasets, along with newly generated free-form grounding instruction-following data. Furthermore, we propose MIG-Bench, a comprehensive benchmark specifically designed for evaluating multi-image grounding capabilities. Experimental results demonstrate that our model achieves significantly superior multi-image grounding capabilities, outperforming the best existing MLLMs by 21.61% and even surpassing much larger 70B models. Our code, model, dataset, and benchmark are fully open-sourced at https://migician-vg.github.io/.
comment: 20 pages, 8 figures
♻ ☆ MusicLIME: Explainable Multimodal Music Understanding ICASSP 2025
Multimodal models are critical for music understanding tasks, as they capture the complex interplay between audio and lyrics. However, as these models become more prevalent, the need for explainability grows-understanding how these systems make decisions is vital for ensuring fairness, reducing bias, and fostering trust. In this paper, we introduce MusicLIME, a model-agnostic feature importance explanation method designed for multimodal music models. Unlike traditional unimodal methods, which analyze each modality separately without considering the interaction between them, often leading to incomplete or misleading explanations, MusicLIME reveals how audio and lyrical features interact and contribute to predictions, providing a holistic view of the model's decision-making. Additionally, we enhance local explanations by aggregating them into global explanations, giving users a broader perspective of model behavior. Through this work, we contribute to improving the interpretability of multimodal music models, empowering users to make informed choices, and fostering more equitable, fair, and transparent music understanding systems.
comment: GitHub repository: https://github.com/IamTheo2000/MusicLIME. To be presented at ICASSP 2025
♻ ☆ Assessment and manipulation of latent constructs in pre-trained language models using psychometric scales
Human-like personality traits have recently been discovered in large language models, raising the hypothesis that their (known and as yet undiscovered) biases conform with human latent psychological constructs. While large conversational models may be tricked into answering psychometric questionnaires, the latent psychological constructs of thousands of simpler transformers, trained for other tasks, cannot be assessed because appropriate psychometric methods are currently lacking. Here, we show how standard psychological questionnaires can be reformulated into natural language inference prompts, and we provide a code library to support the psychometric assessment of arbitrary models. We demonstrate, using a sample of 88 publicly available models, the existence of human-like mental health-related constructs (including anxiety, depression, and Sense of Coherence) which conform with standard theories in human psychology and show similar correlations and mitigation strategies. The ability to interpret and rectify the performance of language models by using psychological tools can boost the development of more explainable, controllable, and trustworthy models.
♻ ☆ InstructOCR: Instruction Boosting Scene Text Spotting AAAI2025
In the field of scene text spotting, previous OCR methods primarily relied on image encoders and pre-trained text information, but they often overlooked the advantages of incorporating human language instructions. To address this gap, we propose InstructOCR, an innovative instruction-based scene text spotting model that leverages human language instructions to enhance the understanding of text within images. Our framework employs both text and image encoders during training and inference, along with instructions meticulously designed based on text attributes. This approach enables the model to interpret text more accurately and flexibly. Extensive experiments demonstrate the effectiveness of our model and we achieve state-of-the-art results on widely used benchmarks. Furthermore, the proposed framework can be seamlessly applied to scene text VQA tasks. By leveraging instruction strategies during pre-training, the performance on downstream VQA tasks can be significantly improved, with a 2.6% increase on the TextVQA dataset and a 2.1% increase on the ST-VQA dataset. These experimental results provide insights into the benefits of incorporating human language instructions for OCR-related tasks.
comment: Accepted by AAAI2025
♻ ☆ II-Bench: An Image Implication Understanding Benchmark for Multimodal Large Language Models
The rapid advancements in the development of multimodal large language models (MLLMs) have consistently led to new breakthroughs on various benchmarks. In response, numerous challenging and comprehensive benchmarks have been proposed to more accurately assess the capabilities of MLLMs. However, there is a dearth of exploration of the higher-order perceptual capabilities of MLLMs. To fill this gap, we propose the Image Implication understanding Benchmark, II-Bench, which aims to evaluate the model's higher-order perception of images. Through extensive experiments on II-Bench across multiple MLLMs, we have made significant findings. Initially, a substantial gap is observed between the performance of MLLMs and humans on II-Bench. The pinnacle accuracy of MLLMs attains 74.8%, whereas human accuracy averages 90%, peaking at an impressive 98%. Subsequently, MLLMs perform worse on abstract and complex images, suggesting limitations in their ability to understand high-level semantics and capture image details. Finally, it is observed that most models exhibit enhanced accuracy when image sentiment polarity hints are incorporated into the prompts. This observation underscores a notable deficiency in their inherent understanding of image sentiment. We believe that II-Bench will inspire the community to develop the next generation of MLLMs, advancing the journey towards expert artificial general intelligence (AGI). II-Bench is publicly available at https://huggingface.co/datasets/m-a-p/II-Bench.
comment: 100 pages, 82 figures, add citations
♻ ☆ Exploring Feature-based Knowledge Distillation for Recommender System: A Frequency Perspective KDD 2025
In this paper, we analyze the feature-based knowledge distillation for recommendation from the frequency perspective. By defining knowledge as different frequency components of the features, we theoretically demonstrate that regular feature-based knowledge distillation is equivalent to equally minimizing losses on all knowledge and further analyze how this equal loss weight allocation method leads to important knowledge being overlooked. In light of this, we propose to emphasize important knowledge by redistributing knowledge weights. Furthermore, we propose FreqD, a lightweight knowledge reweighting method, to avoid the computational cost of calculating losses on each knowledge. Extensive experiments demonstrate that FreqD consistently and significantly outperforms state-of-the-art knowledge distillation methods for recommender systems. Our code is available at https://github.com/woriazzc/KDs.
comment: ACM KDD 2025 Accepted
♻ ☆ WeCromCL: Weakly Supervised Cross-Modality Contrastive Learning for Transcription-only Supervised Text Spotting ECCV 2024
Transcription-only Supervised Text Spotting aims to learn text spotters relying only on transcriptions but no text boundaries for supervision, thus eliminating expensive boundary annotation. The crux of this task lies in locating each transcription in scene text images without location annotations. In this work, we formulate this challenging problem as a Weakly Supervised Cross-modality Contrastive Learning problem, and design a simple yet effective model dubbed WeCromCL that is able to detect each transcription in a scene image in a weakly supervised manner. Unlike typical methods for cross-modality contrastive learning that focus on modeling the holistic semantic correlation between an entire image and a text description, our WeCromCL conducts atomistic contrastive learning to model the character-wise appearance consistency between a text transcription and its correlated region in a scene image to detect an anchor point for the transcription in a weakly supervised manner. The detected anchor points by WeCromCL are further used as pseudo location labels to guide the learning of text spotting. Extensive experiments on four challenging benchmarks demonstrate the superior performance of our model over other methods. Code will be released.
comment: Accepted by ECCV 2024
♻ ☆ AI-Driven Diabetic Retinopathy Screening: Multicentric Validation of AIDRSS in India
Purpose: Diabetic retinopathy (DR) is a major cause of vision loss, particularly in India, where access to retina specialists is limited in rural areas. This study aims to evaluate the Artificial Intelligence-based Diabetic Retinopathy Screening System (AIDRSS) for DR detection and prevalence assessment, addressing the growing need for scalable, automated screening solutions in resource-limited settings. Approach: A multicentric, cross-sectional study was conducted in Kolkata, India, involving 5,029 participants and 10,058 macula-centric retinal fundus images. The AIDRSS employed a deep learning algorithm with 50 million trainable parameters, integrated with Contrast Limited Adaptive Histogram Equalization (CLAHE) preprocessing for enhanced image quality. DR was graded using the International Clinical Diabetic Retinopathy (ICDR) Scale, categorizing disease into five stages (DR0 to DR4). Statistical metrics including sensitivity, specificity, and prevalence rates were evaluated against expert retina specialist assessments. Results: The prevalence of DR in the general population was 13.7%, rising to 38.2% among individuals with elevated random blood glucose levels. The AIDRSS achieved an overall sensitivity of 92%, specificity of 88%, and 100% sensitivity for detecting referable DR (DR3 and DR4). These results demonstrate the system's robust performance in accurately identifying and grading DR in a diverse population. Conclusions: AIDRSS provides a reliable, scalable solution for early DR detection in resource-constrained environments. Its integration of advanced AI techniques ensures high diagnostic accuracy, with potential to significantly reduce the burden of diabetes-related vision loss in underserved regions.
comment: 22 pages, 5 figures. arXiv admin note: substantial text overlap with arXiv:1812.07105 by other authors without attribution
♻ ☆ Quilt-LLaVA: Visual Instruction Tuning by Extracting Localized Narratives from Open-Source Histopathology Videos
Diagnosis in histopathology requires a global whole slide images (WSIs) analysis, requiring pathologists to compound evidence from different WSI patches. The gigapixel scale of WSIs poses a challenge for histopathology multi-modal models. Training multi-model models for histopathology requires instruction tuning datasets, which currently contain information for individual image patches, without a spatial grounding of the concepts within each patch and without a wider view of the WSI. Therefore, they lack sufficient diagnostic capacity for histopathology. To bridge this gap, we introduce Quilt-Instruct, a large-scale dataset of 107,131 histopathology-specific instruction question/answer pairs, grounded within diagnostically relevant image patches that make up the WSI. Our dataset is collected by leveraging educational histopathology videos from YouTube, which provides spatial localization of narrations by automatically extracting the narrators' cursor positions. Quilt-Instruct supports contextual reasoning by extracting diagnosis and supporting facts from the entire WSI. Using Quilt-Instruct, we train Quilt-LLaVA, which can reason beyond the given single image patch, enabling diagnostic reasoning across patches. To evaluate Quilt-LLaVA, we propose a comprehensive evaluation dataset created from 985 images and 1283 human-generated question-answers. We also thoroughly evaluate Quilt-LLaVA using public histopathology datasets, where Quilt-LLaVA significantly outperforms SOTA by over 10% on relative GPT-4 score and 4% and 9% on open and closed set VQA. Our code, data, and model are publicly accessible at quilt-llava.github.io.
♻ ☆ Continual Learning with Strategic Selection and Forgetting for Network Intrusion Detection
Intrusion Detection Systems (IDS) are crucial for safeguarding digital infrastructure. In dynamic network environments, both threat landscapes and normal operational behaviors are constantly changing, resulting in concept drift. While continuous learning mitigates the adverse effects of concept drift, insufficient attention to drift patterns and excessive preservation of outdated knowledge can still hinder the IDS's adaptability. In this paper, we propose SSF (Strategic Selection and Forgetting), a novel continual learning method for IDS, providing continuous model updates with a constantly refreshed memory buffer. Our approach features a strategic sample selection algorithm to select representative new samples and a strategic forgetting mechanism to drop outdated samples. The proposed strategic sample selection algorithm prioritizes new samples that cause the `drifted' pattern, enabling the model to better understand the evolving landscape. Additionally, we introduce strategic forgetting upon detecting significant drift by discarding outdated samples to free up memory, allowing the incorporation of more recent data. SSF captures evolving patterns effectively and ensures the model is aligned with the change of data patterns, significantly enhancing the IDS's adaptability to concept drift. The state-of-the-art performance of SSF on NSL-KDD and UNSW-NB15 datasets demonstrates its superior adaptability to concept drift for network intrusion detection.
comment: Accepted by IEEE International Conference on Computer Communications (INFOCOM) 2025
♻ ☆ MIO: A Foundation Model on Multimodal Tokens
In this paper, we introduce MIO, a novel foundation model built on multimodal tokens, capable of understanding and generating speech, text, images, and videos in an end-to-end, autoregressive manner. While the emergence of large language models (LLMs) and multimodal large language models (MM-LLMs) propels advancements in artificial general intelligence through their versatile capabilities, they still lack true any-to-any understanding and generation. Recently, the release of GPT-4o has showcased the remarkable potential of any-to-any LLMs for complex real-world tasks, enabling omnidirectional input and output across images, speech, and text. However, it is closed-source and does not support the generation of multimodal interleaved sequences. To address this gap, we present MIO, which is trained on a mixture of discrete tokens across four modalities using causal multimodal modeling. MIO undergoes a four-stage training process: (1) alignment pre-training, (2) interleaved pre-training, (3) speech-enhanced pre-training, and (4) comprehensive supervised fine-tuning on diverse textual, visual, and speech tasks. Our experimental results indicate that MIO exhibits competitive, and in some cases superior, performance compared to previous dual-modal baselines, any-to-any model baselines, and even modality-specific baselines. Moreover, MIO demonstrates advanced capabilities inherent to its any-to-any feature, such as interleaved video-text generation, chain-of-visual-thought reasoning, visual guideline generation, instructional image editing, etc.
comment: Technical Report. Codes and models are available in https://github.com/MIO-Team/MIO
♻ ☆ A Comprehensive Study of Structural Pruning for Vision Models
Structural pruning has emerged as a promising approach for producing more efficient models. Nevertheless, the community suffers from a lack of standardized benchmarks and metrics, leaving the progress in this area not fully comprehended.To fill this gap, we present the first comprehensive benchmark, termed PruningBench, for structural pruning. PruningBench showcases the following three characteristics: 1) PruningBench employs a unified and consistent framework for evaluating the effectiveness of diverse structural pruning techniques; 2) PruningBench systematically evaluates 16 existing pruning methods, encompassing a wide array of models (e.g., CNNs and ViTs) and tasks (e.g., classification and detection); 3) PruningBench provides easily implementable interfaces to facilitate the implementation of future pruning methods, and enables the subsequent researchers to incorporate their work into our leaderboards. We provide an online pruning platform for customizing pruning tasks and reproducing all results in this paper. Leaderboard results can also be available.
comment: This is a paper aims to present a evaluation benchmark for structural pruning. The full text is 25 pages
♻ ☆ Buster: Implanting Semantic Backdoor into Text Encoder to Mitigate NSFW Content Generation
The rise of deep learning models in the digital era has raised substantial concerns regarding the generation of Not-Safe-for-Work (NSFW) content. Existing defense methods primarily involve model fine-tuning and post-hoc content moderation. Nevertheless, these approaches largely lack scalability in eliminating harmful content, degrade the quality of benign image generation, or incur high inference costs. To address these challenges, we propose an innovative framework named \textit{Buster}, which injects backdoors into the text encoder to prevent NSFW content generation. Buster leverages deep semantic information rather than explicit prompts as triggers, redirecting NSFW prompts towards targeted benign prompts. Additionally, Buster employs energy-based training data generation through Langevin dynamics for adversarial knowledge augmentation, thereby ensuring robustness in harmful concept definition. This approach demonstrates exceptional resilience and scalability in mitigating NSFW content. Particularly, Buster fine-tunes the text encoder of Text-to-Image models within merely five minutes, showcasing its efficiency. Our extensive experiments denote that Buster outperforms nine state-of-the-art baselines, achieving a superior NSFW content removal rate of at least 91.2\% while preserving the quality of harmless images.
♻ ☆ DiReCT: Diagnostic Reasoning for Clinical Notes via Large Language Models
Large language models (LLMs) have recently showcased remarkable capabilities, spanning a wide range of tasks and applications, including those in the medical domain. Models like GPT-4 excel in medical question answering but may face challenges in the lack of interpretability when handling complex tasks in real clinical settings. We thus introduce the diagnostic reasoning dataset for clinical notes (DiReCT), aiming at evaluating the reasoning ability and interpretability of LLMs compared to human doctors. It contains 511 clinical notes, each meticulously annotated by physicians, detailing the diagnostic reasoning process from observations in a clinical note to the final diagnosis. Additionally, a diagnostic knowledge graph is provided to offer essential knowledge for reasoning, which may not be covered in the training data of existing LLMs. Evaluations of leading LLMs on DiReCT bring out a significant gap between their reasoning ability and that of human doctors, highlighting the critical need for models that can reason effectively in real-world clinical scenarios.
comment: 9 pages,6 figures
♻ ☆ Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability
Mathematical reasoning tasks pose significant challenges for large language models (LLMs) because they require precise logical deduction and sequence analysis. In this work, we introduce the concept of critical tokens -- elements within reasoning trajectories that significantly influence incorrect outcomes. We present a novel framework for identifying these tokens through rollout sampling and demonstrate their substantial divergence from traditional error tokens. Through extensive experiments on datasets such as GSM8K and MATH500, we show that identifying and replacing critical tokens significantly improves model accuracy. We propose an efficient methodology for pinpointing these tokens in large-scale datasets using contrastive estimation and extend this framework to enhance model training processes with direct preference optimization (DPO). Experimental results on GSM8K and MATH500 benchmarks with the widely used models Llama-3 (8B and 70B) and Deepseek-math (7B) demonstrate the effectiveness of the proposed approach, cDPO. Our results underscore the potential of leveraging critical tokens to reduce errors in reasoning tasks, advancing the development of AI systems capable of robust logical deduction. Our code, annotated datasets, and trained models are available at https://github.com/chenzhiling9954/Critical-Tokens-Matter to support and encourage future research in this promising field.
comment: Work in progress
♻ ☆ Large Action Models: From Inception to Implementation
As AI continues to advance, there is a growing demand for systems that go beyond language-based assistance and move toward intelligent agents capable of performing real-world actions. This evolution requires the transition from traditional Large Language Models (LLMs), which excel at generating textual responses, to Large Action Models (LAMs), designed for action generation and execution within dynamic environments. Enabled by agent systems, LAMs hold the potential to transform AI from passive language understanding to active task completion, marking a significant milestone in the progression toward artificial general intelligence. In this paper, we present a comprehensive framework for developing LAMs, offering a systematic approach to their creation, from inception to deployment. We begin with an overview of LAMs, highlighting their unique characteristics and delineating their differences from LLMs. Using a Windows OS-based agent as a case study, we provide a detailed, step-by-step guide on the key stages of LAM development, including data collection, model training, environment integration, grounding, and evaluation. This generalizable workflow can serve as a blueprint for creating functional LAMs in various application domains. We conclude by identifying the current limitations of LAMs and discussing directions for future research and industrial deployment, emphasizing the challenges and opportunities that lie ahead in realizing the full potential of LAMs in real-world applications. The code for the data collection process utilized in this paper is publicly available at: https://github.com/microsoft/UFO/tree/main/dataflow, and comprehensive documentation can be found at https://microsoft.github.io/UFO/dataflow/overview/.
comment: 25pages,12 figures
♻ ☆ Generalizing Weather Forecast to Fine-grained Temporal Scales via Physics-AI Hybrid Modeling
Data-driven artificial intelligence (AI) models have made significant advancements in weather forecasting, particularly in medium-range and nowcasting. However, most data-driven weather forecasting models are black-box systems that focus on learning data mapping rather than fine-grained physical evolution in the time dimension. Consequently, the limitations in the temporal scale of datasets prevent these models from forecasting at finer time scales. This paper proposes a physics-AI hybrid model (i.e., WeatherGFT) which generalizes weather forecasts to finer-grained temporal scales beyond training dataset. Specifically, we employ a carefully designed PDE kernel to simulate physical evolution on a small time scale (e.g., 300 seconds) and use a parallel neural networks with a learnable router for bias correction. Furthermore, we introduce a lead time-aware training framework to promote the generalization of the model at different lead times. The weight analysis of physics-AI modules indicates that physics conducts major evolution while AI performs corrections adaptively. Extensive experiments show that WeatherGFT trained on an hourly dataset, effectively generalizes forecasts across multiple time scales, including 30-minute, which is even smaller than the dataset's temporal resolution.
♻ ☆ Topic-Aware Knowledge Graph with Large Language Models for Interoperability in Recommender Systems
The use of knowledge graphs in recommender systems has become one of the common approaches to addressing data sparsity and cold start problems. Recent advances in large language models (LLMs) offer new possibilities for processing side and context information within knowledge graphs. However, consistent integration across various systems remains challenging due to the need for domain expert intervention and differences in system characteristics. To address these issues, we propose a consistent approach that extracts both general and specific topics from both side and context information using LLMs. First, general topics are iteratively extracted and updated from side information. Then, specific topics are extracted using context information. Finally, to address synonymous topics generated during the specific topic extraction process, a refining algorithm processes and resolves these issues effectively. This approach allows general topics to capture broad knowledge across diverse item characteristics, while specific topics emphasize detailed attributes, providing a more comprehensive understanding of the semantic features of items and the preferences of users. Experimental results demonstrate significant improvements in recommendation performance across diverse knowledge graphs.
comment: Accepted in The 40th ACM/SIGAPP Symposium On Applied Computing(SAC) 2025
♻ ☆ LLM4Vuln: A Unified Evaluation Framework for Decoupling and Enhancing LLMs' Vulnerability Reasoning
Large language models (LLMs) have demonstrated significant potential in various tasks, including those requiring human-level intelligence, such as vulnerability detection. However, recent efforts to use LLMs for vulnerability detection remain preliminary, as they lack a deep understanding of whether a subject LLM's vulnerability reasoning capability stems from the model itself or from external aids such as knowledge retrieval and tooling support. In this paper, we aim to decouple LLMs' vulnerability reasoning from other capabilities, such as vulnerability knowledge adoption, context information retrieval, and advanced prompt schemes. We introduce LLM4Vuln, a unified evaluation framework that separates and assesses LLMs' vulnerability reasoning capabilities and examines improvements when combined with other enhancements. We conduct controlled experiments using 147 ground-truth vulnerabilities and 147 non-vulnerable cases in Solidity, Java and C/C++, testing them in a total of 3,528 scenarios across four LLMs (GPT-3.5, GPT-4, Phi-3, and Llama 3). Our findings reveal the varying impacts of knowledge enhancement, context supplementation, and prompt schemes. We also identify 14 zero-day vulnerabilities in four pilot bug bounty programs, resulting in $3,576 in bounties.
comment: This is a technical report by Nanyang Technological University. Updated to support Solidity, Java and C/C++
♻ ☆ Step-by-Step Mastery: Enhancing Soft Constraint Following Ability of Large Language Models
It is crucial for large language models (LLMs) to follow instructions that involve multiple constraints. However, soft constraints are semantically related and difficult to verify through automated methods. These constraints remain a significant challenge for LLMs. To enhance the ability of LLMs to follow soft constraints, we initially design a pipeline to obtain high-quality outputs automatically. Additionally, to fully utilize the acquired data, we introduce a training paradigm based on curriculum learning. We experimentally evaluate the effectiveness of our methods in improving LLMs' soft constraint following ability and analyze the factors driving the improvements. The datasets and code are publicly available at https://github.com/Rainier-rq/FollowSoftConstraints.
♻ ☆ MLLM-CompBench: A Comparative Reasoning Benchmark for Multimodal LLMs NeurIPS 2024
The ability to compare objects, scenes, or situations is crucial for effective decision-making and problem-solving in everyday life. For instance, comparing the freshness of apples enables better choices during grocery shopping while comparing sofa designs helps optimize the aesthetics of our living space. Despite its significance, the comparative capability is largely unexplored in artificial general intelligence (AGI). In this paper, we introduce MLLM-CompBench, a benchmark designed to evaluate the comparative reasoning capability of multimodal large language models (MLLMs). MLLM-CompBench mines and pairs images through visually oriented questions covering eight dimensions of relative comparison: visual attribute, existence, state, emotion, temporality, spatiality, quantity, and quality. We curate a collection of around 40K image pairs using metadata from diverse vision datasets and CLIP similarity scores. These image pairs span a broad array of visual domains, including animals, fashion, sports, and both outdoor and indoor scenes. The questions are carefully crafted to discern relative characteristics between two images and are labeled by human annotators for accuracy and relevance. We use MLLM-CompBench to evaluate recent MLLMs, including GPT-4V(ision), Gemini-Pro, and LLaVA-1.6. Our results reveal notable shortcomings in their comparative abilities. We believe MLLM-COMPBENCH not only sheds light on these limitations but also establishes a solid foundation for future enhancements in the comparative capability of MLLMs.
comment: This paper has been accepted to NeurIPS 2024. The first two authors contributed equally to this work
♻ ☆ A minimal coalition logic
Coalition Logic is a central logic in logical studies of strategic reasoning, whose models are concurrent game models. In this paper, first, we systematically discuss three assumptions of concurrent game models and argue that they are too strong. The first is seriality; that is, every coalition always has an available joint action. The second is the independence of agents; that is, the merge of two available joint actions of two disjoint coalitions is always an available joint action of the union of the two coalitions. The third is determinism; that is, all available joint actions of the grand coalition always have a unique outcome. Second, we present a coalition logic based on general concurrent game models which do not have the three assumptions and show its completeness. This logic seems minimal for reasoning about coalitional powers.
♻ ☆ HADES: Hardware Accelerated Decoding for Efficient Speculation in Large Language Models
Large Language Models (LLMs) have revolutionized natural language processing by understanding and generating human-like text. However, the increasing demand for more sophisticated LLMs presents significant computational challenges due to their scale and complexity. This paper introduces Hardware Accelerated Decoding (HADES), a novel approach to enhance the performance and energy efficiency of LLMs. We address the design of an LLM accelerator with hardware-level speculative decoding support, a concept not previously explored in existing literature. Our work demonstrates how speculative decoding can significantly improve the efficiency of LLM operations, paving the way for more advanced and practical applications of these models.
comment: Accepted to ICCEA 2025
♻ ☆ Map Imagination Like Blind Humans: Group Diffusion Model for Robotic Map Generation
Can robots imagine or generate maps like humans do, especially when only limited information can be perceived like blind people? To address this challenging task, we propose a novel group diffusion model (GDM) based architecture for robots to generate point cloud maps with very limited input information.Inspired from the blind humans' natural capability of imagining or generating mental maps, the proposed method can generate maps without visual perception data or depth data. With additional limited super-sparse spatial positioning data, like the extra contact-based positioning information the blind individuals can obtain, the map generation quality can be improved even more.Experiments on public datasets are conducted, and the results indicate that our method can generate reasonable maps solely based on path data, and produce even more refined maps upon incorporating exiguous LiDAR data.Compared to conventional mapping approaches, our novel method significantly mitigates sensor dependency, enabling the robots to imagine and generate elementary maps without heavy onboard sensory devices.
♻ ☆ Intelligent System for Automated Molecular Patent Infringement Assessment
Automated drug discovery offers significant potential for accelerating the development of novel therapeutics by substituting labor-intensive human workflows with machine-driven processes. However, molecules generated by artificial intelligence may unintentionally infringe on existing patents, posing legal and financial risks that impede the full automation of drug discovery pipelines. This paper introduces PatentFinder, a novel multi-agent and tool-enhanced intelligence system that can accurately and comprehensively evaluate small molecules for patent infringement. PatentFinder features five specialized agents that collaboratively analyze patent claims and molecular structures with heuristic and model-based tools, generating interpretable infringement reports. To support systematic evaluation, we curate MolPatent-240, a benchmark dataset tailored for patent infringement assessment algorithms. On this benchmark, PatentFinder outperforms baseline methods that rely solely on large language models or specialized chemical tools, achieving a 13.8% improvement in F1-score and a 12% increase in accuracy. Additionally, PatentFinder autonomously generates detailed and interpretable patent infringement reports, showcasing enhanced accuracy and improved interpretability. The high accuracy and interpretability of PatentFinder make it a valuable and reliable tool for automating patent infringement assessments, offering a practical solution for integrating patent protection analysis into the drug discovery pipeline.
♻ ☆ Seeing the Unseen: Learning Basis Confounder Representations for Robust Traffic Prediction KDD 2025
Traffic prediction is essential for intelligent transportation systems and urban computing. It aims to establish a relationship between historical traffic data X and future traffic states Y by employing various statistical or deep learning methods. However, the relations of X -> Y are often influenced by external confounders that simultaneously affect both X and Y , such as weather, accidents, and holidays. Existing deep-learning traffic prediction models adopt the classic front-door and back-door adjustments to address the confounder issue. However, these methods have limitations in addressing continuous or undefined confounders, as they depend on predefined discrete values that are often impractical in complex, real-world scenarios. To overcome this challenge, we propose the Spatial-Temporal sElf-superVised confoundEr learning (STEVE) model. This model introduces a basis vector approach, creating a base confounder bank to represent any confounder as a linear combination of a group of basis vectors. It also incorporates self-supervised auxiliary tasks to enhance the expressive power of the base confounder bank. Afterward, a confounder-irrelevant relation decoupling module is adopted to separate the confounder effects from direct X -> Y relations. Extensive experiments across four large-scale datasets validate our model's superior performance in handling spatial and temporal distribution shifts and underscore its adaptability to unseen confounders. Our model implementation is available at https://github.com/bigscity/STEVE_CODE.
comment: 12 pages, 10 figures, Accepted by KDD 2025
♻ ☆ Explainable Artificial Intelligence: A Survey of Needs, Techniques, Applications, and Future Direction
Artificial intelligence models encounter significant challenges due to their black-box nature, particularly in safety-critical domains such as healthcare, finance, and autonomous vehicles. Explainable Artificial Intelligence (XAI) addresses these challenges by providing explanations for how these models make decisions and predictions, ensuring transparency, accountability, and fairness. Existing studies have examined the fundamental concepts of XAI, its general principles, and the scope of XAI techniques. However, there remains a gap in the literature as there are no comprehensive reviews that delve into the detailed mathematical representations, design methodologies of XAI models, and other associated aspects. This paper provides a comprehensive literature review encompassing common terminologies and definitions, the need for XAI, beneficiaries of XAI, a taxonomy of XAI methods, and the application of XAI methods in different application areas. The survey is aimed at XAI researchers, XAI practitioners, AI model developers, and XAI beneficiaries who are interested in enhancing the trustworthiness, transparency, accountability, and fairness of their AI models.
Robotics 37
☆ SafeSwarm: Decentralized Safe RL for the Swarm of Drones Landing in Dense Crowds
This paper introduces a safe swarm of drones capable of performing landings in crowded environments robustly by relying on Reinforcement Learning techniques combined with Safe Learning. The developed system allows us to teach the swarm of drones with different dynamics to land on moving landing pads in an environment while avoiding collisions with obstacles and between agents. The safe barrier net algorithm was developed and evaluated using a swarm of Crazyflie 2.1 micro quadrotors, which were tested indoors with the Vicon motion capture system to ensure precise localization and control. Experimental results show that our system achieves landing accuracy of 2.25 cm with a mean time of 17 s and collision-free landings, underscoring its effectiveness and robustness in real-world scenarios. This work offers a promising foundation for applications in environments where safety and precision are paramount.
☆ Inductive Learning of Robot Task Knowledge from Raw Data and Online Expert Feedback
The increasing level of autonomy of robots poses challenges of trust and social acceptance, especially in human-robot interaction scenarios. This requires an interpretable implementation of robotic cognitive capabilities, possibly based on formal methods as logics for the definition of task specifications. However, prior knowledge is often unavailable in complex realistic scenarios. In this paper, we propose an offline algorithm based on inductive logic programming from noisy examples to extract task specifications (i.e., action preconditions, constraints and effects) directly from raw data of few heterogeneous (i.e., not repetitive) robotic executions. Our algorithm leverages on the output of any unsupervised action identification algorithm from video-kinematic recordings. Combining it with the definition of very basic, almost task-agnostic, commonsense concepts about the environment, which contribute to the interpretability of our methodology, we are able to learn logical axioms encoding preconditions of actions, as well as their effects in the event calculus paradigm. Since the quality of learned specifications depends mainly on the accuracy of the action identification algorithm, we also propose an online framework for incremental refinement of task knowledge from user feedback, guaranteeing safe execution. Results in a standard manipulation task and benchmark for user training in the safety-critical surgical robotic scenario, show the robustness, data- and time-efficiency of our methodology, with promising results towards the scalability in more complex domains.
☆ The Sense of Agency in Assistive Robotics Using Shared Autonomy
Sense of agency is one factor that influences people's preferences for robot assistance and a phenomenon from cognitive science that represents the experience of control over one's environment. However, in assistive robotics literature, we often see paradigms that optimize measures like task success and cognitive load, rather than sense of agency. In fact, prior work has found that participants sometimes express a preference for paradigms, such as direct teleoperation, which do not perform well with those other metrics but give more control to the user. In this work, we focus on a subset of assistance paradigms for manipulation called shared autonomy in which the system combines control signals from the user and the automated control. We run a study to evaluate sense of agency and show that higher robot autonomy during assistance leads to improved task performance but a decreased sense of agency, indicating a potential trade-off between task performance and sense of agency. From our findings, we discuss the relation between sense of agency and optimality, and we consider a proxy metric for a component of sense of agency which might enable us to build systems that monitor and maintain sense of agency in real time.
comment: 10 pages, 8 figure, HRI conference
☆ Empirical Comparison of Four Stereoscopic Depth Sensing Cameras for Robotics Applications
Depth sensing is an essential technology in robotics and many other fields. Many depth sensing (or RGB-D) cameras are available on the market and selecting the best one for your application can be challenging. In this work, we tested four stereoscopic RGB-D cameras that sense the distance by using two images from slightly different views. We empirically compared four cameras (Intel RealSense D435, Intel RealSense D455, StereoLabs ZED 2, and Luxonis OAK-D Pro) in three scenarios: (i) planar surface perception, (ii) plastic doll perception, (iii) household object perception (YCB dataset). We recorded and evaluated more than 3,000 RGB-D frames for each camera. For table-top robotics scenarios with distance to objects up to one meter, the best performance is provided by the D435 camera. For longer distances, the other three models perform better, making them more suitable for some mobile robotics applications. OAK-D Pro additionally offers integrated AI modules (e.g., object and human keypoint detection). ZED 2 is not a standalone device and requires a computer with a GPU for depth data acquisition. All data (more than 12,000 RGB-D frames) are publicly available at https://osf.io/f2seb.
☆ Efficiently Closing Loops in LiDAR-Based SLAM Using Point Cloud Density Maps
Consistent maps are key for most autonomous mobile robots. They often use SLAM approaches to build such maps. Loop closures via place recognition help maintain accurate pose estimates by mitigating global drift. This paper presents a robust loop closure detection pipeline for outdoor SLAM with LiDAR-equipped robots. The method handles various LiDAR sensors with different scanning patterns, field of views and resolutions. It generates local maps from LiDAR scans and aligns them using a ground alignment module to handle both planar and non-planar motion of the LiDAR, ensuring applicability across platforms. The method uses density-preserving bird's eye view projections of these local maps and extracts ORB feature descriptors from them for place recognition. It stores the feature descriptors in a binary search tree for efficient retrieval, and self-similarity pruning addresses perceptual aliasing in repetitive environments. Extensive experiments on public and self-recorded datasets demonstrate accurate loop closure detection, long-term localization, and cross-platform multi-map alignment, agnostic to the LiDAR scanning patterns, fields of view, and motion profiles.
☆ Fast-Revisit Coverage Path Planning for Autonomous Mobile Patrol Robots Using Long-Range Sensor Information
The utilization of Unmanned Ground Vehicles (UGVs) for patrolling industrial sites has expanded significantly. These UGVs typically are equipped with perception systems, e.g., computer vision, with limited range due to sensor limitations or site topology. High-level control of the UGVs requires Coverage Path Planning (CPP) algorithms that navigate all relevant waypoints and promptly start the next cycle. In this paper, we propose the novel Fast-Revisit Coverage Path Planning (FaRe-CPP) algorithm using a greedy heuristic approach to propose waypoints for maximum coverage area and a random search-based path optimization technique to obtain a path along the proposed waypoints with minimum revisit time. We evaluated the algorithm in a simulated environment using Gazebo and a camera-equipped TurtleBot3 against a number of existing algorithms. Compared to their average revisit times and path lengths, our FaRe-CPP algorithm approximately showed a 45% and 40% reduction, respectively, in these highly relevant performance indicators.
☆ Evaluation of Artificial Intelligence Methods for Lead Time Prediction in Non-Cycled Areas of Automotive Production
The present study examines the effectiveness of applying Artificial Intelligence methods in an automotive production environment to predict unknown lead times in a non-cycle-controlled production area. Data structures are analyzed to identify contextual features and then preprocessed using one-hot encoding. Methods selection focuses on supervised machine learning techniques. In supervised learning methods, regression and classification methods are evaluated. Continuous regression based on target size distribution is not feasible. Classification methods analysis shows that Ensemble Learning and Support Vector Machines are the most suitable. Preliminary study results indicate that gradient boosting algorithms LightGBM, XGBoost, and CatBoost yield the best results. After further testing and extensive hyperparameter optimization, the final method choice is the LightGBM algorithm. Depending on feature availability and prediction interval granularity, relative prediction accuracies of up to 90% can be achieved. Further tests highlight the importance of periodic retraining of AI models to accurately represent complex production processes using the database. The research demonstrates that AI methods can be effectively applied to highly variable production data, adding business value by providing an additional metric for various control tasks while outperforming current non AI-based systems.
comment: 7 pages, 4 figures, CLC2024 Conference
☆ ViewVR: Visual Feedback Modes to Achieve Quality of VR-based Telemanipulation
The paper focuses on an immersive teleoperation system that enhances operator's ability to actively perceive the robot's surroundings. A consumer-grade HTC Vive VR system was used to synchronize the operator's hand and head movements with a UR3 robot and a custom-built robotic head with two degrees of freedom (2-DoF). The system's usability, manipulation efficiency, and intuitiveness of control were evaluated in comparison with static head camera positioning across three distinct tasks. Code and other supplementary materials can be accessed by link: https://github.com/ErkhovArtem/ViewVR
☆ GestLLM: Advanced Hand Gesture Interpretation via Large Language Models for Human-Robot Interaction
This paper introduces GestLLM, an advanced system for human-robot interaction that enables intuitive robot control through hand gestures. Unlike conventional systems, which rely on a limited set of predefined gestures, GestLLM leverages large language models and feature extraction via MediaPipe to interpret a diverse range of gestures. This integration addresses key limitations in existing systems, such as restricted gesture flexibility and the inability to recognize complex or unconventional gestures commonly used in human communication. By combining state-of-the-art feature extraction and language model capabilities, GestLLM achieves performance comparable to leading vision-language models while supporting gestures underrepresented in traditional datasets. For example, this includes gestures from popular culture, such as the ``Vulcan salute" from Star Trek, without any additional pretraining, prompt engineering, etc. This flexibility enhances the naturalness and inclusivity of robot control, making interactions more intuitive and user-friendly. GestLLM provides a significant step forward in gesture-based interaction, enabling robots to understand and respond to a wide variety of hand gestures effectively. This paper outlines its design, implementation, and evaluation, demonstrating its potential applications in advanced human-robot collaboration, assistive robotics, and interactive entertainment.
☆ PO-GVINS: Tightly Coupled GNSS-Visual-Inertial Integration with Pose-Only Representation
Accurate and reliable positioning is crucial for perception, decision-making, and other high-level applications in autonomous driving, unmanned aerial vehicles, and intelligent robots. Given the inherent limitations of standalone sensors, integrating heterogeneous sensors with complementary capabilities is one of the most effective approaches to achieving this goal. In this paper, we propose a filtering-based, tightly coupled global navigation satellite system (GNSS)-visual-inertial positioning framework with a pose-only formulation applied to the visual-inertial system (VINS), termed PO-GVINS. Specifically, multiple-view imaging used in current VINS requires a priori of 3D feature, then jointly estimate camera poses and 3D feature position, which inevitably introduces linearization error of the feature as well as facing dimensional explosion. However, the pose-only (PO) formulation, which is demonstrated to be equivalent to the multiple-view imaging and has been applied in visual reconstruction, represent feature depth using two camera poses and thus 3D feature position is removed from state vector avoiding aforementioned difficulties. Inspired by this, we first apply PO formulation in our VINS, i.e., PO-VINS. GNSS raw measurements are then incorporated with integer ambiguity resolved to achieve accurate and drift-free estimation. Extensive experiments demonstrate that the proposed PO-VINS significantly outperforms the multi-state constrained Kalman filter (MSCKF). By incorporating GNSS measurements, PO-GVINS achieves accurate, drift-free state estimation, making it a robust solution for positioning in challenging environments.
☆ GazeGrasp: DNN-Driven Robotic Grasping with Wearable Eye-Gaze Interface
We present GazeGrasp, a gaze-based manipulation system enabling individuals with motor impairments to control collaborative robots using eye-gaze. The system employs an ESP32 CAM for eye tracking, MediaPipe for gaze detection, and YOLOv8 for object localization, integrated with a Universal Robot UR10 for manipulation tasks. After user-specific calibration, the system allows intuitive object selection with a magnetic snapping effect and robot control via eye gestures. Experimental evaluation involving 13 participants demonstrated that the magnetic snapping effect significantly reduced gaze alignment time, improving task efficiency by 31%. GazeGrasp provides a robust, hands-free interface for assistive robotics, enhancing accessibility and autonomy for users.
comment: Accepted to: IEEE/ACM International Conference on Human-Robot Interaction (HRI 2025)
☆ Touched by ChatGPT: Using an LLM to Drive Affective Tactile Interaction
Touch is a fundamental aspect of emotion-rich communication, playing a vital role in human interaction and offering significant potential in human-robot interaction. Previous research has demonstrated that a sparse representation of human touch can effectively convey social tactile signals. However, advances in human-robot tactile interaction remain limited, as many humanoid robots possess simplistic capabilities, such as only opening and closing their hands, restricting nuanced tactile expressions. In this study, we explore how a robot can use sparse representations of tactile vibrations to convey emotions to a person. To achieve this, we developed a wearable sleeve integrated with a 5x5 grid of vibration motors, enabling the robot to communicate diverse tactile emotions and gestures. Using chain prompts within a Large Language Model (LLM), we generated distinct 10-second vibration patterns corresponding to 10 emotions (e.g., happiness, sadness, fear) and 6 touch gestures (e.g., pat, rub, tap). Participants (N = 32) then rated each vibration stimulus based on perceived valence and arousal. People are accurate at recognising intended emotions, a result which aligns with earlier findings. These results highlight the LLM's ability to generate emotional haptic data and effectively convey emotions through tactile signals. By translating complex emotional and tactile expressions into vibratory patterns, this research demonstrates how LLMs can enhance physical interaction between humans and robots.
☆ Improving Incremental Nonlinear Dynamic Inversion Robustness Using Robust Control in Aerial Robotics
Improving robustness to uncertainty and rejection of external disturbances represents a significant challenge in aerial robotics. Nonlinear controllers based on Incremental Nonlinear Dynamic Inversion (INDI), known for their ability in estimating disturbances through measured-filtered data, have been notably used in such applications. Typically, these controllers comprise two cascaded loops: an inner loop employing nonlinear dynamic inversion and an outer loop generating the virtual control inputs via linear controllers. In this paper, a novel methodology is introduced, that combines the advantages of INDI with the robustness of linear structured $\mathcal{H}_\infty$ controllers. A full cascaded architecture is proposed to control the dynamics of a multirotor drone, covering both stabilization and guidance. In particular, low-order $\mathcal{H}_\infty$ controllers are designed for the outer loop by properly structuring the problem and solving it through non-smooth optimization. A comparative analysis is conducted between an existing INDI/PD approach and the proposed INDI/$\mathcal{H}_\infty$ strategy, showing a notable enhancement in the rejection of external disturbances. It is carried out first using MATLAB simulations involving a nonlinear model of a Parrot Bebop quadcopter drone, and then experimentally using a customized quadcopter built by the ENAC team. The results show an improvement of more than 50\% in the rejection of disturbances such as gusts.
☆ Temperature Driven Multi-modal/Single-actuated Soft Finger
Soft pneumatic fingers are of great research interest. However, their significant potential is limited as most of them can generate only one motion, mostly bending. The conventional design of soft fingers does not allow them to switch to another motion mode. In this paper, we developed a novel multi-modal and single-actuated soft finger where its motion mode is switched by changing the finger's temperature. Our soft finger is capable of switching between three distinctive motion modes: bending, twisting, and extension-in approximately five seconds. We carried out a detailed experimental study of the soft finger and evaluated its repeatability and range of motion. It exhibited repeatability of around one millimeter and a fifty percent larger range of motion than a standard bending actuator. We developed an analytical model for a fiber-reinforced soft actuator for twisting motion. This helped us relate the input pressure to the output twist radius of the twisting motion. This model was validated by experimental verification. Further, a soft robotic gripper with multiple grasp modes was developed using three actuators. This gripper can adapt to and grasp objects of a large range of size, shape, and stiffness. We showcased its grasping capabilities by successfully grasping a small berry, a large roll, and a delicate tofu cube.
☆ Multi-face emotion detection for effective Human-Robot Interaction
The integration of dialogue interfaces in mobile devices has become ubiquitous, providing a wide array of services. As technology progresses, humanoid robots designed with human-like features to interact effectively with people are gaining prominence, and the use of advanced human-robot dialogue interfaces is continually expanding. In this context, emotion recognition plays a crucial role in enhancing human-robot interaction by enabling robots to understand human intentions. This research proposes a facial emotion detection interface integrated into a mobile humanoid robot, capable of displaying real-time emotions from multiple individuals on a user interface. To this end, various deep neural network models for facial expression recognition were developed and evaluated under consistent computer-based conditions, yielding promising results. Afterwards, a trade-off between accuracy and memory footprint was carefully considered to effectively implement this application on a mobile humanoid robot.
comment: 9 pages, 8 figures and 1 table. Accepted at the 17th International Conference on Agents and Artificial Intelligence (ICAART 2025), Porto, Portugal
☆ Evaluating Robotic Approach Techniques for the Insertion of a Straight Instrument into a Vitreoretinal Surgery Trocar
Advances in vitreoretinal robotic surgery enable precise techniques for gene therapies. This study evaluates three robotic approaches using the 7-DoF robotic arm for docking a micro-precise tool to a trocar: fully co-manipulated, hybrid co-manipulated/teleoperated, and hybrid with camera assistance. The fully co-manipulated approach was the fastest but had a 42% success rate. Hybrid methods showed higher success rates (91.6% and 100%) and completed tasks within 2 minutes. NASA Task Load Index (TLX) assessments indicated lower physical demand and effort for hybrid approaches.
comment: 2 Pages, 2 Figures, 1 Table
☆ ROSAnnotator: A Web Application for ROSBag Data Analysis in Human-Robot Interaction
Human-robot interaction (HRI) is an interdisciplinary field that utilises both quantitative and qualitative methods. While ROSBags, a file format within the Robot Operating System (ROS), offer an efficient means of collecting temporally synched multimodal data in empirical studies with real robots, there is a lack of tools specifically designed to integrate qualitative coding and analysis functions with ROSBags. To address this gap, we developed ROSAnnotator, a web-based application that incorporates a multimodal Large Language Model (LLM) to support both manual and automated annotation of ROSBag data. ROSAnnotator currently facilitates video, audio, and transcription annotations and provides an open interface for custom ROS messages and tools. By using ROSAnnotator, researchers can streamline the qualitative analysis process, create a more cohesive analysis pipeline, and quickly access statistical summaries of annotations, thereby enhancing the overall efficiency of HRI data analysis. https://github.com/CHRI-Lab/ROSAnnotator
comment: Accepted to HRI 2025
☆ Sthymuli: a Static Educational Robot. Leveraging the Thymio II Platform ICRA40
The use of robots in education represents a challenge for teachers and a fixed vision of what robots can do for students. This paper presents the development of Sthymuli, a static educational robot designed to explore new classroom interactions between robots, students and teachers. We propose the use of the Thymio II educational platform as a base, ensuring a robust benchmark for a fair comparison of the commonly available wheeled robots and our exploratory approach with Sthymuli. This paper outlines the constraints and requirements for developing such a robot, the current state of development and future work.
comment: Two pages, three figures. ICRA40 extended abstract
☆ Motion Tracks: A Unified Representation for Human-Robot Transfer in Few-Shot Imitation Learning
Teaching robots to autonomously complete everyday tasks remains a challenge. Imitation Learning (IL) is a powerful approach that imbues robots with skills via demonstrations, but is limited by the labor-intensive process of collecting teleoperated robot data. Human videos offer a scalable alternative, but it remains difficult to directly train IL policies from them due to the lack of robot action labels. To address this, we propose to represent actions as short-horizon 2D trajectories on an image. These actions, or motion tracks, capture the predicted direction of motion for either human hands or robot end-effectors. We instantiate an IL policy called Motion Track Policy (MT-pi) which receives image observations and outputs motion tracks as actions. By leveraging this unified, cross-embodiment action space, MT-pi completes tasks with high success given just minutes of human video and limited additional robot demonstrations. At test time, we predict motion tracks from two camera views, recovering 6DoF trajectories via multi-view synthesis. MT-pi achieves an average success rate of 86.5% across 4 real-world tasks, outperforming state-of-the-art IL baselines which do not leverage human data or our action space by 40%, and generalizes to scenarios seen only in human videos. Code and videos are available on our website https://portal-cornell.github.io/motion_track_policy/.
☆ Hand-Object Contact Detection using Grasp Quality Metrics
We propose a novel hand-object contact detection system based on grasp quality metrics extracted from object and hand poses, and evaluated its performance using the DexYCB dataset. Our evaluation demonstrated the system's high accuracy (approaching 90%). Future work will focus on a real-time implementation using vision-based estimation, and integrating it to a robot-to-human handover system.
comment: Submitted to the 2025 IEEE/ACM International Conference on Human-Robot Interaction (HRI'25)
☆ Testing Human-Hand Segmentation on In-Distribution and Out-of-Distribution Data in Human-Robot Interactions Using a Deep Ensemble Model
Reliable detection and segmentation of human hands are critical for enhancing safety and facilitating advanced interactions in human-robot collaboration. Current research predominantly evaluates hand segmentation under in-distribution (ID) data, which reflects the training data of deep learning (DL) models. However, this approach fails to address out-of-distribution (OOD) scenarios that often arise in real-world human-robot interactions. In this study, we present a novel approach by evaluating the performance of pre-trained DL models under both ID data and more challenging OOD scenarios. To mimic realistic industrial scenarios, we designed a diverse dataset featuring simple and cluttered backgrounds with industrial tools, varying numbers of hands (0 to 4), and hands with and without gloves. For OOD scenarios, we incorporated unique and rare conditions such as finger-crossing gestures and motion blur from fast-moving hands, addressing both epistemic and aleatoric uncertainties. To ensure multiple point of views (PoVs), we utilized both egocentric cameras, mounted on the operator's head, and static cameras to capture RGB images of human-robot interactions. This approach allowed us to account for multiple camera perspectives while also evaluating the performance of models trained on existing egocentric datasets as well as static-camera datasets. For segmentation, we used a deep ensemble model composed of UNet and RefineNet as base learners. Performance evaluation was conducted using segmentation metrics and uncertainty quantification via predictive entropy. Results revealed that models trained on industrial datasets outperformed those trained on non-industrial datasets, highlighting the importance of context-specific training. Although all models struggled with OOD scenarios, those trained on industrial datasets demonstrated significantly better generalization.
☆ Autonomous Electrochemistry Platform with Real-Time Normality Testing of Voltammetry Measurements Using ML
Electrochemistry workflows utilize various instruments and computing systems to execute workflows consisting of electrocatalyst synthesis, testing and evaluation tasks. The heterogeneity of the software and hardware of these ecosystems makes it challenging to orchestrate a complete workflow from production to characterization by automating its tasks. We propose an autonomous electrochemistry computing platform for a multi-site ecosystem that provides the services for remote experiment steering, real-time measurement transfer, and AI/ML-driven analytics. We describe the integration of a mobile robot and synthesis workstation into the ecosystem by developing custom hub-networks and software modules to support remote operations over the ecosystem's wireless and wired networks. We describe a workflow task for generating I-V voltammetry measurements using a potentiostat, and a machine learning framework to ensure their normality by detecting abnormal conditions such as disconnected electrodes. We study a number of machine learning methods for the underlying detection problem, including smooth, non-smooth, structural and statistical methods, and their fusers. We present experimental results to illustrate the effectiveness of this platform, and also validate the proposed ML method by deriving its rigorous generalization equations.
comment: 10 pages, 14 figures, accepted in the IEEE 20th International Conference on e-Science (e-Science), 2024
♻ ☆ Few-Shot Task Learning through Inverse Generative Modeling
Learning the intents of an agent, defined by its goals or motion style, is often extremely challenging from just a few examples. We refer to this problem as task concept learning and present our approach, Few-Shot Task Learning through Inverse Generative Modeling (FTL-IGM), which learns new task concepts by leveraging invertible neural generative models. The core idea is to pretrain a generative model on a set of basic concepts and their demonstrations. Then, given a few demonstrations of a new concept (such as a new goal or a new action), our method learns the underlying concepts through backpropagation without updating the model weights, thanks to the invertibility of the generative model. We evaluate our method in five domains -- object rearrangement, goal-oriented navigation, motion caption of human actions, autonomous driving, and real-world table-top manipulation. Our experimental results demonstrate that via the pretrained generative model, we successfully learn novel concepts and generate agent plans or motion corresponding to these concepts in (1) unseen environments and (2) in composition with training concepts.
comment: Added acknowledgment
♻ ☆ Accelerating genetic optimization of nonlinear model predictive control by learning optimal search space size
Genetic algorithm (GA) is typically used to solve nonlinear model predictive control's optimization problem. However, the size of the search space in which the GA searches for the optimal control inputs is crucial for its applicability to fast-response systems. This paper proposes accelerating the genetic optimization of NMPC by learning optimal search space size. The approach trains a multivariate regression model to adaptively predict the best smallest size of the search space in every control cycle. The proposed approach reduces the GA's computational time, improves the chance of convergence to better control inputs, and provides a stable and feasible solution. The proposed approach was evaluated on three nonlinear systems and compared to four other evolutionary algorithms implemented in a processor-in-the-loop fashion. The results show that the proposed approach provides a 17-45\% reduction in computational time and increases the convergence rate by 35-47\%. The source code is available on GitHub.
comment: Accepted by the Journal of Control and Decision
♻ ☆ Geometric Freeze-Tag Problem
We study the Freeze-Tag Problem (FTP), introduced by Arkin et al. (SODA'02), where the objective is to activate a group of n robots, starting from a single initially active robot. Robots are positioned in $\mathbb{R}^d$, and once activated, they move at a constant speed to wake up others. The goal is to minimize the time required to activate the last robot, known as the makespan. We establish new upper bounds for the makespan under the $l_1$ and $l_2$ norms in $\mathbb{R}^2$ and $\mathbb{R}^3$. Specifically, we improve the previous upper bound for $(\mathbb{R}^2, l_2)$ from $7.07r$ (Bonichon et al., DISC'24) to $5.064r$. For $(\mathbb{R}^3, l_1)$, we derive a makespan bound of $13r$, which translates to $22.52r$ for $(\mathbb{R}^3, l_2)$. Here, $r$ denotes the maximum distance of any robot from the initially active robot under the given norm. To our knowledge, these are the first makespan bounds for FTP in $\mathbb{R}^3$. Additionally, we show that the maximum makespan for $n$ robots is not necessarily achieved when robots are equally distributed along the boundary in $(\mathbb{R}^2, l_2)$. We further investigate FTP in $(\mathbb{R}^3, l_2)$ for specific configurations where robots lie on a boundary, providing insights into practical scenarios.
♻ ☆ QuadWBG: Generalizable Quadrupedal Whole-Body Grasping
Legged robots with advanced manipulation capabilities have the potential to significantly improve household duties and urban maintenance. Despite considerable progress in developing robust locomotion and precise manipulation methods, seamlessly integrating these into cohesive whole-body control for real-world applications remains challenging. In this paper, we present a modular framework for robust and generalizable whole-body loco-manipulation controller based on a single arm-mounted camera. By using reinforcement learning (RL), we enable a robust low-level policy for command execution over 5 dimensions (5D) and a grasp-aware high-level policy guided by a novel metric, Generalized Oriented Reachability Map (GORM). The proposed system achieves state-of-the-art one-time grasping accuracy of 89% in the real world, including challenging tasks such as grasping transparent objects. Through extensive simulations and real-world experiments, we demonstrate that our system can effectively manage a large workspace, from floor level to above body height, and perform diverse whole-body loco-manipulation tasks.
♻ ☆ SyncDiff: Synchronized Motion Diffusion for Multi-Body Human-Object Interaction Synthesis
Synthesizing realistic human-object interaction motions is a critical problem in VR/AR and human animation. Unlike the commonly studied scenarios involving a single human or hand interacting with one object, we address a more generic multi-body setting with arbitrary numbers of humans, hands, and objects. This complexity introduces significant challenges in synchronizing motions due to the high correlations and mutual influences among bodies. To address these challenges, we introduce SyncDiff, a novel method for multi-body interaction synthesis using a synchronized motion diffusion strategy. SyncDiff employs a single diffusion model to capture the joint distribution of multi-body motions. To enhance motion fidelity, we propose a frequency-domain motion decomposition scheme. Additionally, we introduce a new set of alignment scores to emphasize the synchronization of different body motions. SyncDiff jointly optimizes both data sample likelihood and alignment likelihood through an explicit synchronization strategy. Extensive experiments across four datasets with various multi-body configurations demonstrate the superiority of SyncDiff over existing state-of-the-art motion synthesis methods.
♻ ☆ An Adaptive Sliding Window Estimator for Positioning of Unmanned Aerial Vehicle Using a Single Anchor
Localization using a single range anchor combined with onboard optical-inertial odometry offers a lightweight solution that provides multidimensional measurements for the positioning of unmanned aerial vehicles. Unfortunately, the performance of such lightweight sensors varies with the dynamic environment, and the fidelity of the dynamic model is also severely affected by environmental aerial flow. To address this challenge, we propose an adaptive sliding window estimator equipped with an estimation reliability evaluator, where the states, noise covariance matrices and aerial drag are estimated simultaneously. The aerial drag effects are first evaluated based on posterior states and covariance. Then, an augmented Kalman filter is designed to pre-process multidimensional measurements and inherit historical information. Subsequently, an inverse-Wishart smoother is employed to estimate posterior states and covariance matrices. To further suppress potential divergence, a reliability evaluator is devised to infer estimation errors. We further determine the fidelity of each sensor based on the error propagation. Extensive experiments are conducted in both standard and harsh environments, demonstrating the adaptability and robustness of the proposed method. The root mean square error reaches 0.15 m, outperforming the state-of-the-art approach.
comment: This work has been submitted to the IEEE for possible publication
♻ ☆ Walk along: An Experiment on Controlling the Mobile Robot 'Spot' with Voice and Gestures
Robots are becoming more capable and can autonomously perform tasks such as navigating between locations. However, human oversight remains crucial. This study compared two touchless methods for directing mobile robots: voice control and gesture control, to investigate the efficiency of the methods and the preference of users. We tested these methods in two conditions: one in which participants remained stationary and one in which they walked freely alongside the robot. We hypothesized that walking alongside the robot would result in higher intuitiveness ratings and improved task performance, based on the idea that walking promotes spatial alignment and reduces the effort required for mental rotation. In a 2x2 within-subject design, 218 participants guided the quadruped robot Spot along a circuitous route with multiple 90-degree turns using rotate left, rotate right, and walk forward commands. After each trial, participants rated the intuitiveness of the command mapping, while post-experiment interviews were used to gather the participants' preferences. Results showed that voice control combined with walking with Spot was the most favored and intuitive, whereas gesture control while standing caused confusion for left/right commands. Nevertheless, 29% of participants preferred gesture control, citing increased task engagement and visual congruence as reasons. An odometry-based analysis revealed that participants often followed behind Spot, particularly in the gesture control condition, when they were allowed to walk. In conclusion, voice control with walking produced the best outcomes. Improving physical ergonomics and adjusting gesture types could make gesture control more effective.
♻ ☆ Adaptive Non-linear Centroidal MPC with Stability Guarantees for Robust Locomotion of Legged Robots
Nonlinear model predictive locomotion controllers based on the reduced centroidal dynamics are nowadays ubiquitous in legged robots. These schemes, even if they assume an inherent simplification of the robot's dynamics, were shown to endow robots with a step-adjustment capability in reaction to small pushes, and, moreover, in the case of uncertain parameters - as unknown payloads - they were shown to be able to provide some practical, albeit limited, robustness. In this work, we provide rigorous certificates of their closed loop stability via a reformulation of the centroidal MPC controller. This is achieved thanks to a systematic procedure inspired by the machinery of adaptive control, together with ideas coming from Control Lyapunov functions. Our reformulation, in addition, provides robustness for a class of unmeasured constant disturbances. To demonstrate the generality of our approach, we validated our formulation on a new generation of humanoid robots - the 56.7 kg ergoCub, as well as on a commercially available 21 kg quadruped robot, Aliengo.
♻ ☆ From Underground Mines to Offices: A Versatile and Robust Framework for Range-Inertial SLAM
Simultaneous Localization and Mapping (SLAM) is an essential component of autonomous robotic applications and self-driving vehicles, enabling them to understand and operate in their environment. Many SLAM systems have been proposed in the last decade, but they are often complex to adapt to different settings or sensor setups. In this work, we present LiDAR Graph-SLAM (LG-SLAM), a versatile range-inertial SLAM framework that can be adapted to different types of sensors and environments, from underground mines to offices with minimal parameter tuning. Our system integrates range, inertial and GNSS measurements into a graph-based optimization framework. We also use a refined submap management approach and a robust loop closure method that effectively accounts for uncertainty in the identification and validation of putative loop closures, ensuring global consistency and robustness. Enabled by a parallelized architecture and GPU integration, our system achieves pose estimation at LiDAR frame rate, along with online loop closing and graph optimization. We validate our system in diverse environments using public datasets and real-world data, consistently achieving an average error below 20 cm and outperforming other state-of-the-art algorithms.
comment: 8 pages, 8 figures, 3 tables
♻ ☆ LLaMAR: Long-Horizon Planning for Multi-Agent Robots in Partially Observable Environments
The ability of Language Models (LMs) to understand natural language makes them a powerful tool for parsing human instructions into task plans for autonomous robots. Unlike traditional planning methods that rely on domain-specific knowledge and handcrafted rules, LMs generalize from diverse data and adapt to various tasks with minimal tuning, acting as a compressed knowledge base. However, LMs in their standard form face challenges with long-horizon tasks, particularly in partially observable multi-agent settings. We propose an LM-based Long-Horizon Planner for Multi-Agent Robotics (LLaMAR), a cognitive architecture for planning that achieves state-of-the-art results in long-horizon tasks within partially observable environments. LLaMAR employs a plan-act-correct-verify framework, allowing self-correction from action execution feedback without relying on oracles or simulators. Additionally, we present MAP-THOR, a comprehensive test suite encompassing household tasks of varying complexity within the AI2-THOR environment. Experiments show that LLaMAR achieves a 30% higher success rate than other state-of-the-art LM-based multi-agent planners in MAP-THOR and Search \& Rescue tasks. Code can be found at https://github.com/nsidn98/LLaMAR
comment: 27 pages, 4 figures, 5 tables
♻ ☆ Map Imagination Like Blind Humans: Group Diffusion Model for Robotic Map Generation
Can robots imagine or generate maps like humans do, especially when only limited information can be perceived like blind people? To address this challenging task, we propose a novel group diffusion model (GDM) based architecture for robots to generate point cloud maps with very limited input information.Inspired from the blind humans' natural capability of imagining or generating mental maps, the proposed method can generate maps without visual perception data or depth data. With additional limited super-sparse spatial positioning data, like the extra contact-based positioning information the blind individuals can obtain, the map generation quality can be improved even more.Experiments on public datasets are conducted, and the results indicate that our method can generate reasonable maps solely based on path data, and produce even more refined maps upon incorporating exiguous LiDAR data.Compared to conventional mapping approaches, our novel method significantly mitigates sensor dependency, enabling the robots to imagine and generate elementary maps without heavy onboard sensory devices.
♻ ☆ Robot Error Awareness Through Human Reactions: Implementation, Evaluation, and Recommendations
Effective error detection is crucial to prevent task disruption and maintain user trust. Traditional methods often rely on task-specific models or user reporting, which can be inflexible or slow. Recent research suggests social signals, naturally exhibited by users in response to robot errors, can enable more flexible, timely error detection. However, most studies rely on post hoc analysis, leaving their real-time effectiveness uncertain and lacking user-centric evaluation. In this work, we developed a proactive error detection system that combines user behavioral signals (facial action units and speech), user feedback, and error context for automatic error detection. In a study (N = 28), we compared our proactive system to a status quo reactive approach. Results show our system 1) reliably and flexibly detects error, 2) detects errors faster than the reactive approach, and 3) is perceived more favorably by users than the reactive one. We discuss recommendations for enabling robot error awareness in future HRI systems.
♻ ☆ Efficient Estimation of Relaxed Model Parameters for Robust UAV Trajectory Optimization
Online trajectory optimization and optimal control methods are crucial for enabling sustainable unmanned aerial vehicle (UAV) services, such as agriculture, environmental monitoring, and transportation, where available actuation and energy are limited. However, optimal controllers are highly sensitive to model mismatch, which can occur due to loaded equipment, packages to be delivered, or pre-existing variability in fundamental structural and thrust-related parameters. To circumvent this problem, optimal controllers can be paired with parameter estimators to improve their trajectory planning performance and perform adaptive control. However, UAV platforms are limited in terms of onboard processing power, oftentimes making nonlinear parameter estimation too computationally expensive to consider. To address these issues, we propose a relaxed, affine-in-parameters multirotor model along with an efficient optimal parameter estimator. We convexify the nominal Moving Horizon Parameter Estimation (MHPE) problem into a linear-quadratic form (LQ-MHPE) via an affine-in-parameter relaxation on the nonlinear dynamics, resulting in fast quadratic programs (QPs) that facilitate adaptive Model Predictve Control (MPC) in real time. We compare this approach to the equivalent nonlinear estimator in Monte Carlo simulations, demonstrating a decrease in average solve time and trajectory optimality cost by 98.2% and 23.9-56.2%, respectively.
comment: 8 pages, 5 figures, to be published in IEEE Sustech 2025
♻ ☆ A Mixed-Integer Conic Program for the Moving-Target Traveling Salesman Problem based on a Graph of Convex Sets
This paper introduces a new formulation that finds the optimum for the Moving-Target Traveling Salesman Problem (MT-TSP), which seeks to find a shortest path for an agent, that starts at a depot, visits a set of moving targets exactly once within their assigned time-windows, and returns to the depot. The formulation relies on the key idea that when the targets move along lines, their trajectories become convex sets within the space-time coordinate system. The problem then reduces to finding the shortest path within a graph of convex sets, subject to some speed constraints. We compare our formulation with the current state-of-the-art Mixed Integer Conic Program (MICP) solver for the MT-TSP. The experimental results show that our formulation outperforms the MICP for instances with up to 20 targets, with up to two orders of magnitude reduction in runtime, and up to a 60\% tighter optimality gap. We also show that the solution cost from the convex relaxation of our formulation provides significantly tighter lower bounds for the MT-TSP than the ones from the MICP.
comment: 7 pages, 4 figures
♻ ☆ Exploiting Chordal Sparsity for Fast Global Optimality with Application to Localization SP
In recent years, many estimation problems in robotics have been shown to be solvable to global optimality using their semidefinite relaxations. However, the runtime complexity of off-the-shelf semidefinite programming (SDP) solvers is up to cubic in problem size, which inhibits real-time solutions of problems involving large state dimensions. We show that for a large class of problems, namely those with chordal sparsity, we can reduce the complexity of these solvers to linear in problem size. In particular, we show how to replace the large positive-semidefinite variable with a number of smaller interconnected ones using the well-known chordal decomposition. This formulation also allows for the straightforward application of the alternating direction method of multipliers (ADMM), which can exploit parallelism for increased scalability. We show for two example problems in simulation that the chordal solvers provide a significant speed-up over standard SDP solvers, and that global optimality is crucial in the absence of good initializations.
comment: 21 pages, 6 figures. Version history: v1: initial arXiv, v2: WAFR submission, v3: correction, v4: WAFR conference-ready, v5: WAFR SPAR journal version
Multimedia 7
☆ Audio-CoT: Exploring Chain-of-Thought Reasoning in Large Audio Language Model
Large Audio-Language Models (LALMs) have demonstrated remarkable performance in tasks involving audio perception and understanding, such as speech recognition and audio captioning. However, their reasoning capabilities - critical for solving complex real-world problems - remain underexplored. In this work, we conduct the first exploration into integrating Chain-of-Thought (CoT) reasoning into LALMs to enhance their reasoning ability across auditory modalities. We evaluate representative CoT methods, analyzing their performance in both information extraction and reasoning tasks across sound, music, and speech domains. Our findings reveal that CoT methods significantly improve performance on easy and medium tasks but encounter challenges with hard tasks, where reasoning chains can confuse the model rather than improve accuracy. Additionally, we identify a positive correlation between reasoning path length and accuracy, demonstrating the potential of scaling inference for advanced instruction-following and reasoning. This study not only highlights the promise of CoT in enhancing LALM reasoning capabilities but also identifies key limitations and provides actionable directions for future research.
☆ Depth and Image Fusion for Road Obstacle Detection Using Stereo Camera
This paper is devoted to the detection of objects on a road, performed with a combination of two methods based on both the use of depth information and video analysis of data from a stereo camera. Since neither the time of the appearance of an object on the road, nor its size and shape is known in advance, ML/DL-based approaches are not applicable. The task becomes more complicated due to variations in artificial illumination, inhomogeneous road surface texture, and unknown character and features of the object. To solve this problem we developed the depth and image fusion method that complements a search of small contrast objects by RGB-based method, and obstacle detection by stereo image-based approach with SLIC superpixel segmentation. We conducted experiments with static and low speed obstacles in an underground parking lot and demonstrated the successful work of the developed technique for detecting and even tracking small objects, which can be parking infrastructure objects, things left on the road, wheels, dropped boxes, etc.
comment: 8 pages, 15 figures
☆ Dynamic Multimodal Fusion via Meta-Learning Towards Micro-Video Recommendation
Multimodal information (e.g., visual, acoustic, and textual) has been widely used to enhance representation learning for micro-video recommendation. For integrating multimodal information into a joint representation of micro-video, multimodal fusion plays a vital role in the existing micro-video recommendation approaches. However, the static multimodal fusion used in previous studies is insufficient to model the various relationships among multimodal information of different micro-videos. In this paper, we develop a novel meta-learning-based multimodal fusion framework called Meta Multimodal Fusion (MetaMMF), which dynamically assigns parameters to the multimodal fusion function for each micro-video during its representation learning. Specifically, MetaMMF regards the multimodal fusion of each micro-video as an independent task. Based on the meta information extracted from the multimodal features of the input task, MetaMMF parameterizes a neural network as the item-specific fusion function via a meta learner. We perform extensive experiments on three benchmark datasets, demonstrating the significant improvements over several state-of-the-art multimodal recommendation models, like MMGCN, LATTICE, and InvRL. Furthermore, we lighten our model by adopting canonical polyadic decomposition to improve the training efficiency, and validate its effectiveness through experimental results. Codes are available at https://github.com/hanliu95/MetaMMF.
comment: This paper has been accepted by ACM Transactions on Information Systems
☆ Pedestrian Trajectory Prediction Based on Social Interactions Learning With Random Weights
Pedestrian trajectory prediction is a critical technology in the evolution of self-driving cars toward complete artificial intelligence. Over recent years, focusing on the trajectories of pedestrians to model their social interactions has surged with great interest in more accurate trajectory predictions. However, existing methods for modeling pedestrian social interactions rely on pre-defined rules, struggling to capture non-explicit social interactions. In this work, we propose a novel framework named DTGAN, which extends the application of Generative Adversarial Networks (GANs) to graph sequence data, with the primary objective of automatically capturing implicit social interactions and achieving precise predictions of pedestrian trajectory. DTGAN innovatively incorporates random weights within each graph to eliminate the need for pre-defined interaction rules. We further enhance the performance of DTGAN by exploring diverse task loss functions during adversarial training, which yields improvements of 16.7\% and 39.3\% on metrics ADE and FDE, respectively. The effectiveness and accuracy of our framework are verified on two public datasets. The experimental results show that our proposed DTGAN achieves superior performance and is well able to understand pedestrians' intentions.
comment: 13 pages,7 figures,Accepted to IEEE Transactions on Multimedia (TMM)
☆ An Efficient NVoD Scheme Using Implicit Error Correction and Subchannels for Wireless Networks
Implicit Error Correction (IEC) is a near Video-on-Demand (nVoD) scheme that trades bandwidth utilization for initial playback delay to potentially support an infinite number of users. Additionally, it provides error protection without any further bandwidth increase by exploiting the implicit redundancy of nVoD protocols, using linear combinations of the segments transmitted in a given time slot. However, IEC packet loss protection is weaker at the beginning of the playback due to the lack of implicit redundancy and lower decoding efficiency, resulting in worse subjective playback quality. In tackling this issue, this paper contributes with an extension of the original nVoD architecture, enhancing its performance by adding a new element namely, subchannels. These subdivisions of the original channels do not provide further packet loss protection but significantly improve the decoding efficiency, which in turn increases playback quality, especially at the beginning. Even for very high packet loss probabilities, subchannels are designed to obtain higher decoding efficiency which results in greater packet loss protection than that provided by IEC. The proposed scheme is especially useful in wireless cooperative networks using techniques such as network coding, as content transmissions can be split into different subchannels in order to maximize network efficiency.
♻ ☆ The Sound of Water: Inferring Physical Properties from Pouring Liquids ICASSP 2025
We study the connection between audio-visual observations and the underlying physics of a mundane yet intriguing everyday activity: pouring liquids. Given only the sound of liquid pouring into a container, our objective is to automatically infer physical properties such as the liquid level, the shape and size of the container, the pouring rate and the time to fill. To this end, we: (i) show in theory that these properties can be determined from the fundamental frequency (pitch); (ii) train a pitch detection model with supervision from simulated data and visual data with a physics-inspired objective; (iii) introduce a new large dataset of real pouring videos for a systematic study; (iv) show that the trained model can indeed infer these physical properties for real data; and finally, (v) we demonstrate strong generalization to various container shapes, other datasets, and in-the-wild YouTube videos. Our work presents a keen understanding of a narrow yet rich problem at the intersection of acoustics, physics, and learning. It opens up applications to enhance multisensory perception in robotic pouring.
comment: Project page at https://bpiyush.github.io/pouring-water-website. Short version accepted to ICASSP 2025
♻ ☆ BeFA: A General Behavior-driven Feature Adapter for Multimedia Recommendation AAAI2025
Multimedia recommender systems focus on utilizing behavioral information and content information to model user preferences. Typically, it employs pre-trained feature encoders to extract content features, then fuses them with behavioral features. However, pre-trained feature encoders often extract features from the entire content simultaneously, including excessive preference-irrelevant details. We speculate that it may result in the extracted features not containing sufficient features to accurately reflect user preferences. To verify our hypothesis, we introduce an attribution analysis method for visually and intuitively analyzing the content features. The results indicate that certain products' content features exhibit the issues of information drift}and information omission,reducing the expressive ability of features. Building upon this finding, we propose an effective and efficient general Behavior-driven Feature Adapter (BeFA) to tackle these issues. This adapter reconstructs the content feature with the guidance of behavioral information, enabling content features accurately reflecting user preferences. Extensive experiments demonstrate the effectiveness of the adapter across all multimedia recommendation methods. Our code is made publicly available on https://github.com/fqldom/BeFA.
comment: This paper is accepted by AAAI2025
Computer Vision and Pattern Recognition 24
☆ Comparison of Autoencoders for tokenization of ASL datasets
Generative AI, powered by large language models (LLMs), has revolutionized applications across text, audio, images, and video. This study focuses on developing and evaluating encoder-decoder architectures for the American Sign Language (ASL) image dataset, consisting of 87,000 images across 29 hand sign classes. Three approaches were compared: Feedforward Autoencoders, Convolutional Autoencoders, and Diffusion Autoencoders. The Diffusion Autoencoder outperformed the others, achieving the lowest mean squared error (MSE) and highest Mean Opinion Score (MOS) due to its probabilistic noise modeling and iterative denoising capabilities. The Convolutional Autoencoder demonstrated effective spatial feature extraction but lacked the robustness of the diffusion process, while the Feedforward Autoencoder served as a baseline with limitations in handling complex image data. Objective and subjective evaluations confirmed the superiority of the Diffusion Autoencoder for high-fidelity image reconstruction, emphasizing its potential in multimodal AI applications such as sign language recognition and generation. This work provides critical insights into designing robust encoder-decoder systems to advance multimodal AI capabilities.
comment: 9 pages, 2 tables, 4 figures
☆ Super-Resolution of 3D Micro-CT Images Using Generative Adversarial Networks: Enhancing Resolution and Segmentation Accuracy
We develop a procedure for substantially improving the quality of segmented 3D micro-Computed Tomography (micro-CT) images of rocks with a Machine Learning (ML) Generative Model. The proposed model enhances the resolution eightfold (8x) and addresses segmentation inaccuracies due to the overlapping X-ray attenuation in micro-CT measurement for different rock minerals and phases. The proposed generative model is a 3D Deep Convolutional Wasserstein Generative Adversarial Network with Gradient Penalty (3D DC WGAN-GP). The algorithm is trained on segmented 3D low-resolution micro-CT images and segmented unpaired complementary 2D high-resolution Laser Scanning Microscope (LSM) images. The algorithm was demonstrated on multiple samples of Berea sandstones. We achieved high-quality super-resolved 3D images with a resolution of 0.4375 micro-m/voxel and accurate segmentation for constituting minerals and pore space. The described procedure can significantly expand the modern capabilities of digital rock physics.
comment: 24 pages, 9 figures
☆ Evaluating unsupervised contrastive learning framework for MRI sequences classification
The automatic identification of Magnetic Resonance Imaging (MRI) sequences can streamline clinical workflows by reducing the time radiologists spend manually sorting and identifying sequences, thereby enabling faster diagnosis and treatment planning for patients. However, the lack of standardization in the parameters of MRI scans poses challenges for automated systems and complicates the generation and utilization of datasets for machine learning research. To address this issue, we propose a system for MRI sequence identification using an unsupervised contrastive deep learning framework. By training a convolutional neural network based on the ResNet-18 architecture, our system classifies nine common MRI sequence types as a 9-class classification problem. The network was trained using an in-house internal dataset and validated on several public datasets, including BraTS, ADNI, Fused Radiology-Pathology Prostate Dataset, the Breast Cancer Dataset (ACRIN), among others, encompassing diverse acquisition protocols and requiring only 2D slices for training. Our system achieves a classification accuracy of over 0.95 across the nine most common MRI sequence types.
☆ CULTURE3D: Cultural Landmarks and Terrain Dataset for 3D Applications
In this paper, we present a large-scale fine-grained dataset using high-resolution images captured from locations worldwide. Compared to existing datasets, our dataset offers a significantly larger size and includes a higher level of detail, making it uniquely suited for fine-grained 3D applications. Notably, our dataset is built using drone-captured aerial imagery, which provides a more accurate perspective for capturing real-world site layouts and architectural structures. By reconstructing environments with these detailed images, our dataset supports applications such as the COLMAP format for Gaussian Splatting and the Structure-from-Motion (SfM) method. It is compatible with widely-used techniques including SLAM, Multi-View Stereo, and Neural Radiance Fields (NeRF), enabling accurate 3D reconstructions and point clouds. This makes it a benchmark for reconstruction and segmentation tasks. The dataset enables seamless integration with multi-modal data, supporting a range of 3D applications, from architectural reconstruction to virtual tourism. Its flexibility promotes innovation, facilitating breakthroughs in 3D modeling and analysis.
☆ Benchmarking YOLOv8 for Optimal Crack Detection in Civil Infrastructure
Ensuring the structural integrity and safety of bridges is crucial for the reliability of transportation networks and public safety. Traditional crack detection methods are increasingly being supplemented or replaced by advanced artificial intelligence (AI) techniques. However, most of the models rely on two-stage target detection algorithms, which pose concerns for real-time applications due to their lower speed. While models such as YOLO (You Only Look Once) have emerged as transformative tools due to their remarkable speed and accuracy. However, the potential of the latest YOLOv8 framework in this domain remains underexplored. This study bridges that gap by rigorously evaluating YOLOv8's performance across five model scales (nano, small, medium, large, and extra-large) using a high-quality Roboflow dataset. A comprehensive hyperparameter optimization was performed, testing six state-of-the-art optimizers-Stochastic Gradient Descent, Adaptive Moment Estimation, Adam with Decoupled Weight Decay, Root Mean Square Propagation, Rectified Adam, and Nesterov-accelerated Adam. Results revealed that YOLOv8, optimized with Stochastic Gradient Descent, delivered exceptional accuracy and speed, setting a new benchmark for real-time crack detection. Beyond its immediate application, this research positions YOLOv8 as a foundational approach for integrating advanced computer vision techniques into infrastructure monitoring. By enabling more reliable and proactive maintenance of aging bridge networks, this work paves the way for safer, more efficient transportation systems worldwide.
comment: Accepted at 104th TRB Annual Meeting 2025
☆ Driver Age and Its Effect on Key Driving Metrics: Insights from Dynamic Vehicle Data
By 2030, the senior population aged 65 and older is expected to increase by over 50%, significantly raising the number of older drivers on the road. Drivers over 70 face higher crash death rates compared to those in their forties and fifties, underscoring the importance of developing more effective safety interventions for this demographic. Although the impact of aging on driving behavior has been studied, there is limited research on how these behaviors translate into real-world driving scenarios. This study addresses this need by leveraging Naturalistic Driving Data (NDD) to analyze driving performance measures - specifically, speed limit adherence on interstates and deceleration at stop intersections, both of which may be influenced by age-related declines. Using NDD, we developed Cumulative Distribution Functions (CDFs) to establish benchmarks for key driving behaviors among senior and young drivers. Our analysis, which included anomaly detection, benchmark comparisons, and accuracy evaluations, revealed significant differences in driving patterns primarily related to speed limit adherence at 75mph. While our approach shows promising potential for enhancing Advanced Driver Assistance Systems (ADAS) by providing tailored interventions based on age-specific adherence to speed limit driving patterns, we recognize the need for additional data to refine and validate metrics for other driving behaviors. By establishing precise benchmarks for various driving performance metrics, ADAS can effectively identify anomalies, such as abrupt deceleration, which may indicate impaired driving or other safety concerns. This study lays a strong foundation for future research aimed at improving safety interventions through detailed driving behavior analysis.
comment: 21 pages, 9 figures, 4 Tables, 104th TRB Annual Meeting 2025, Washington DC
☆ Local Foreground Selection aware Attentive Feature Reconstruction for few-shot fine-grained plant species classification
Plant species exhibit significant intra-class variation and minimal inter-class variation. To enhance classification accuracy, it is essential to reduce intra-class variation while maximizing inter-class variation. This paper addresses plant species classification using a limited number of labelled samples and introduces a novel Local Foreground Selection(LFS) attention mechanism. LFS is a straightforward module designed to generate discriminative support and query feature maps. It operates by integrating two types of attention: local attention, which captures local spatial details to enhance feature discrimination and increase inter-class differentiation, and foreground selection attention, which emphasizes the foreground plant object while mitigating background interference. By focusing on the foreground, the query and support features selectively highlight relevant feature sequences and disregard less significant background sequences, thereby reducing intra-class differences. Experimental results from three plant species datasets demonstrate the effectiveness of the proposed LFS attention mechanism and its complementary advantages over previous feature reconstruction methods.
☆ Synthetic Prior for Few-Shot Drivable Head Avatar Inversion
We present SynShot, a novel method for the few-shot inversion of a drivable head avatar based on a synthetic prior. We tackle two major challenges. First, training a controllable 3D generative network requires a large number of diverse sequences, for which pairs of images and high-quality tracked meshes are not always available. Second, state-of-the-art monocular avatar models struggle to generalize to new views and expressions, lacking a strong prior and often overfitting to a specific viewpoint distribution. Inspired by machine learning models trained solely on synthetic data, we propose a method that learns a prior model from a large dataset of synthetic heads with diverse identities, expressions, and viewpoints. With few input images, SynShot fine-tunes the pretrained synthetic prior to bridge the domain gap, modeling a photorealistic head avatar that generalizes to novel expressions and viewpoints. We model the head avatar using 3D Gaussian splatting and a convolutional encoder-decoder that outputs Gaussian parameters in UV texture space. To account for the different modeling complexities over parts of the head (e.g., skin vs hair), we embed the prior with explicit control for upsampling the number of per-part primitives. Compared to state-of-the-art monocular methods that require thousands of real training images, SynShot significantly improves novel view and expression synthesis.
comment: Website https://zielon.github.io/synshot/
☆ ActiveGAMER: Active GAussian Mapping through Efficient Rendering
We introduce ActiveGAMER, an active mapping system that utilizes 3D Gaussian Splatting (3DGS) to achieve high-quality, real-time scene mapping and exploration. Unlike traditional NeRF-based methods, which are computationally demanding and restrict active mapping performance, our approach leverages the efficient rendering capabilities of 3DGS, allowing effective and efficient exploration in complex environments. The core of our system is a rendering-based information gain module that dynamically identifies the most informative viewpoints for next-best-view planning, enhancing both geometric and photometric reconstruction accuracy. ActiveGAMER also integrates a carefully balanced framework, combining coarse-to-fine exploration, post-refinement, and a global-local keyframe selection strategy to maximize reconstruction completeness and fidelity. Our system autonomously explores and reconstructs environments with state-of-the-art geometric and photometric accuracy and completeness, significantly surpassing existing approaches in both aspects. Extensive evaluations on benchmark datasets such as Replica and MP3D highlight ActiveGAMER's effectiveness in active mapping tasks.
☆ MedGrad E-CLIP: Enhancing Trust and Transparency in AI-Driven Skin Lesion Diagnosis WACV
As deep learning models gain attraction in medical data, ensuring transparent and trustworthy decision-making is essential. In skin cancer diagnosis, while advancements in lesion detection and classification have improved accuracy, the black-box nature of these methods poses challenges in understanding their decision processes, leading to trust issues among physicians. This study leverages the CLIP (Contrastive Language-Image Pretraining) model, trained on different skin lesion datasets, to capture meaningful relationships between visual features and diagnostic criteria terms. To further enhance transparency, we propose a method called MedGrad E-CLIP, which builds on gradient-based E-CLIP by incorporating a weighted entropy mechanism designed for complex medical imaging like skin lesions. This approach highlights critical image regions linked to specific diagnostic descriptions. The developed integrated pipeline not only classifies skin lesions by matching corresponding descriptions but also adds an essential layer of explainability developed especially for medical data. By visually explaining how different features in an image relates to diagnostic criteria, this approach demonstrates the potential of advanced vision-language models in medical image analysis, ultimately improving transparency, robustness, and trust in AI-driven diagnostic systems.
comment: Accepted to 2025 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW)
☆ Transforming Vision Transformer: Towards Efficient Multi-Task Asynchronous Learning NeurIPS 2024
Multi-Task Learning (MTL) for Vision Transformer aims at enhancing the model capability by tackling multiple tasks simultaneously. Most recent works have predominantly focused on designing Mixture-of-Experts (MoE) structures and in tegrating Low-Rank Adaptation (LoRA) to efficiently perform multi-task learning. However, their rigid combination hampers both the optimization of MoE and the ef fectiveness of reparameterization of LoRA, leading to sub-optimal performance and low inference speed. In this work, we propose a novel approach dubbed Efficient Multi-Task Learning (EMTAL) by transforming a pre-trained Vision Transformer into an efficient multi-task learner during training, and reparameterizing the learned structure for efficient inference. Specifically, we firstly develop the MoEfied LoRA structure, which decomposes the pre-trained Transformer into a low-rank MoE structure and employ LoRA to fine-tune the parameters. Subsequently, we take into account the intrinsic asynchronous nature of multi-task learning and devise a learning Quality Retaining (QR) optimization mechanism, by leveraging the historical high-quality class logits to prevent a well-trained task from performance degradation. Finally, we design a router fading strategy to integrate the learned parameters into the original Transformer, archiving efficient inference. Extensive experiments on public benchmarks demonstrate the superiority of our method, compared to the state-of-the-art multi-task learning approaches.
comment: Accepted by the 38th Conference on Neural Information Processing Systems (NeurIPS 2024)
☆ Real-Time Neural-Enhancement for Online Cloud Gaming
Online Cloud gaming demands real-time, high-quality video transmission across variable wide-area networks (WANs). Neural-enhanced video transmission algorithms employing super-resolution (SR) for video quality enhancement have effectively challenged WAN environments. However, these SR-based methods require intensive fine-tuning for the whole video, making it infeasible in diverse online cloud gaming. To address this, we introduce River, a cloud gaming delivery framework designed based on the observation that video segment features in cloud gaming are typically repetitive and redundant. This permits a significant opportunity to reuse fine-tuned SR models, reducing the fine-tuning latency of minutes to query latency of milliseconds. To enable the idea, we design a practical system that addresses several challenges, such as model organization, online model scheduler, and transfer strategy. River first builds a content-aware encoder that fine-tunes SR models for diverse video segments and stores them in a lookup table. When delivering cloud gaming video streams online, River checks the video features and retrieves the most relevant SR models to enhance the frame quality. Meanwhile, if no existing SR model performs well enough for some video segments, River will further fine-tune new models and update the lookup table. Finally, to avoid the overhead of streaming model weight to the clients, River designs a prefetching strategy that predicts the models with the highest possibility of being retrieved. Our evaluation based on real video game streaming demonstrates River can reduce redundant training overhead by 44% and improve the Peak-Signal-to-Noise-Ratio by 1.81dB compared to the SOTA solutions. Practical deployment shows River meets real-time requirements, achieving approximately 720p 20fps on mobile devices.
☆ Defect Detection Network In PCB Circuit Devices Based on GAN Enhanced YOLOv11
This study proposes an advanced method for surface defect detection in printed circuit boards (PCBs) using an improved YOLOv11 model enhanced with a generative adversarial network (GAN). The approach focuses on identifying six common defect types: missing hole, rat bite, open circuit, short circuit, burr, and virtual welding. By employing GAN to generate synthetic defect images, the dataset is augmented with diverse and realistic patterns, improving the model's ability to generalize, particularly for complex and infrequent defects like burrs. The enhanced YOLOv11 model is evaluated on a PCB defect dataset, demonstrating significant improvements in accuracy, recall, and robustness, especially when dealing with defects in complex environments or small targets. This research contributes to the broader field of electronic design automation (EDA), where efficient defect detection is a crucial step in ensuring high-quality PCB manufacturing. By integrating advanced deep learning techniques, this approach enhances the automation and precision of defect detection, reducing reliance on manual inspection and accelerating design-to-production workflows. The findings underscore the importance of incorporating GAN-based data augmentation and optimized detection architectures in EDA processes, providing valuable insights for improving reliability and efficiency in PCB defect detection within industrial applications.
☆ Uncertainty-Aware Online Extrinsic Calibration: A Conformal Prediction Approach WACV 2025
Accurate sensor calibration is crucial for autonomous systems, yet its uncertainty quantification remains underexplored. We present the first approach to integrate uncertainty awareness into online extrinsic calibration, combining Monte Carlo Dropout with Conformal Prediction to generate prediction intervals with a guaranteed level of coverage. Our method proposes a framework to enhance existing calibration models with uncertainty quantification, compatible with various network architectures. Validated on KITTI (RGB Camera-LiDAR) and DSEC (Event Camera-LiDAR) datasets, we demonstrate effectiveness across different visual sensor types, measuring performance with adapted metrics to evaluate the efficiency and reliability of the intervals. By providing calibration parameters with quantifiable confidence measures, we offer insights into the reliability of calibration estimates, which can greatly improve the robustness of sensor fusion in dynamic environments and usefully serve the Computer Vision community.
comment: Accepted for publication at WACV 2025
☆ A Foundational Generative Model for Breast Ultrasound Image Analysis
Foundational models have emerged as powerful tools for addressing various tasks in clinical settings. However, their potential development to breast ultrasound analysis remains untapped. In this paper, we present BUSGen, the first foundational generative model specifically designed for breast ultrasound image analysis. Pretrained on over 3.5 million breast ultrasound images, BUSGen has acquired extensive knowledge of breast structures, pathological features, and clinical variations. With few-shot adaptation, BUSGen can generate repositories of realistic and informative task-specific data, facilitating the development of models for a wide range of downstream tasks. Extensive experiments highlight BUSGen's exceptional adaptability, significantly exceeding real-data-trained foundational models in breast cancer screening, diagnosis, and prognosis. In breast cancer early diagnosis, our approach outperformed all board-certified radiologists (n=9), achieving an average sensitivity improvement of 16.5% (P-value<0.0001). Additionally, we characterized the scaling effect of using generated data which was as effective as the collected real-world data for training diagnostic models. Moreover, extensive experiments demonstrated that our approach improved the generalization ability of downstream models. Importantly, BUSGen protected patient privacy by enabling fully de-identified data sharing, making progress forward in secure medical data utilization. An online demo of BUSGen is available at https://aibus.bio.
comment: Peking University; Stanford University; Peking University Cancer Hospital & Institute; Peking Union Medical College Hospital; Cancer Hospital, Chinese Academy of Medical Sciences
☆ LarvSeg: Exploring Image Classification Data For Large Vocabulary Semantic Segmentation via Category-wise Attentive Classifier
Scaling up the vocabulary of semantic segmentation models is extremely challenging because annotating large-scale mask labels is labour-intensive and time-consuming. Recently, language-guided segmentation models have been proposed to address this challenge. However, their performance drops significantly when applied to out-of-distribution categories. In this paper, we propose a new large vocabulary semantic segmentation framework, called LarvSeg. Different from previous works, LarvSeg leverages image classification data to scale the vocabulary of semantic segmentation models as large-vocabulary classification datasets usually contain balanced categories and are much easier to obtain. However, for classification tasks, the category is image-level, while for segmentation we need to predict the label at pixel level. To address this issue, we first propose a general baseline framework to incorporate image-level supervision into the training process of a pixel-level segmentation model, making the trained network perform semantic segmentation on newly introduced categories in the classification data. We then observe that a model trained on segmentation data can group pixel features of categories beyond the training vocabulary. Inspired by this finding, we design a category-wise attentive classifier to apply supervision to the precise regions of corresponding categories to improve the model performance. Extensive experiments demonstrate that LarvSeg significantly improves the large vocabulary semantic segmentation performance, especially in the categories without mask labels. For the first time, we provide a 21K-category semantic segmentation model with the help of ImageNet21K. The code is available at https://github.com/HaojunYu1998/large_voc_seg.
comment: PRCV 2024
☆ A General Framework for Inference-time Scaling and Steering of Diffusion Models
Diffusion models produce impressive results in modalities ranging from images and video to protein design and text. However, generating samples with user-specified properties remains a challenge. Recent research proposes fine-tuning models to maximize rewards that capture desired properties, but these methods require expensive training and are prone to mode collapse. In this work, we propose Feynman Kac (FK) steering, an inference-time framework for steering diffusion models with reward functions. FK steering works by sampling a system of multiple interacting diffusion processes, called particles, and resampling particles at intermediate steps based on scores computed using functions called potentials. Potentials are defined using rewards for intermediate states and are selected such that a high value indicates that the particle will yield a high-reward sample. We explore various choices of potentials, intermediate rewards, and samplers. We evaluate FK steering on text-to-image and text diffusion models. For steering text-to-image models with a human preference reward, we find that FK steering a 0.8B parameter model outperforms a 2.6B parameter fine-tuned model on prompt fidelity, with faster sampling and no training. For steering text diffusion models with rewards for text quality and specific text attributes, we find that FK steering generates lower perplexity, more linguistically acceptable outputs and enables gradient-free control of attributes like toxicity. Our results demonstrate that inference-time scaling and steering of diffusion models, even with off-the-shelf rewards, can provide significant sample quality gains and controllability benefits. Code is available at https://github.com/zacharyhorvitz/Fk-Diffusion-Steering .
☆ Faithful Counterfactual Visual Explanations (FCVE)
Deep learning models in computer vision have made remarkable progress, but their lack of transparency and interpretability remains a challenge. The development of explainable AI can enhance the understanding and performance of these models. However, existing techniques often struggle to provide convincing explanations that non-experts easily understand, and they cannot accurately identify models' intrinsic decision-making processes. To address these challenges, we propose to develop a counterfactual explanation (CE) model that balances plausibility and faithfulness. This model generates easy-to-understand visual explanations by making minimum changes necessary in images without altering the pixel data. Instead, the proposed method identifies internal concepts and filters learned by models and leverages them to produce plausible counterfactual explanations. The provided explanations reflect the internal decision-making process of the model, thus ensuring faithfulness to the model.
♻ ☆ DoubleDiffusion: Combining Heat Diffusion with Denoising Diffusion for Generative Learning on 3D Meshes
This paper proposes DoubleDiffusion, a novel framework that combines heat dissipation diffusion and denoising diffusion for direct generative learning on 3D mesh surfaces. Our approach addresses the challenges of generating continuous signal distributions residing on a curve manifold surface. Unlike previous methods that rely on unrolling 3D meshes into 2D or adopting field representations, DoubleDiffusion leverages the Laplacian-Beltrami operator to process features respecting the mesh structure. This combination enables effective geometry-aware signal diffusion across the underlying geometry. As shown in Fig.1, we demonstrate that DoubleDiffusion has the ability to generate RGB signal distributions on complex 3D mesh surfaces and achieves per-category shape-conditioned texture generation across different shape geometry. Our work contributes a new direction in diffusion-based generative modeling on 3D surfaces, with potential applications in the field of 3D asset generation.
comment: Codes: https://github.com/Wxyxixixi/DoubleDiffusion_3D_Mesh
♻ ☆ Artificial Intelligence for Cochlear Implants: Review of Strategies, Challenges, and Perspectives
Automatic speech recognition (ASR) plays a pivotal role in our daily lives, offering utility not only for interacting with machines but also for facilitating communication for individuals with partial or profound hearing impairments. The process involves receiving the speech signal in analog form, followed by various signal processing algorithms to make it compatible with devices of limited capacities, such as cochlear implants (CIs). Unfortunately, these implants, equipped with a finite number of electrodes, often result in speech distortion during synthesis. Despite efforts by researchers to enhance received speech quality using various state-of-the-art (SOTA) signal processing techniques, challenges persist, especially in scenarios involving multiple sources of speech, environmental noise, and other adverse conditions. The advent of new artificial intelligence (AI) methods has ushered in cutting-edge strategies to address the limitations and difficulties associated with traditional signal processing techniques dedicated to CIs. This review aims to comprehensively cover advancements in CI-based ASR and speech enhancement, among other related aspects. The primary objective is to provide a thorough overview of metrics and datasets, exploring the capabilities of AI algorithms in this biomedical field, and summarizing and commenting on the best results obtained. Additionally, the review will delve into potential applications and suggest future directions to bridge existing research gaps in this domain.
♻ ☆ Exploring Superpixel Segmentation Methods in the Context of Citizen Science and Deforestation Detection
Tropical forests play an essential role in the planet's ecosystem, making the conservation of these biomes a worldwide priority. However, ongoing deforestation and degradation pose a significant threat to their existence, necessitating effective monitoring and the proposal of actions to mitigate the damage caused by these processes. In this regard, initiatives range from government and private sector monitoring programs to solutions based on citizen science campaigns, for example. Particularly in the context of citizen science campaigns, the segmentation of remote sensing images to identify deforested areas and subsequently submit them to analysis by non-specialized volunteers is necessary. Thus, segmentation using superpixel-based techniques proves to be a viable solution for this important task. Therefore, this paper presents an analysis of 22 superpixel-based segmentation methods applied to remote sensing images, aiming to identify which of them are more suitable for generating segments for citizen science campaigns. The results reveal that seven of the segmentation methods outperformed the baseline method (SLIC) currently employed in the ForestEyes citizen science project, indicating an opportunity for improvement in this important stage of campaign development.
comment: This paper is under review
♻ ☆ Fresh-CL: Feature Realignment through Experts on Hypersphere in Continual Learning ICASSP 2025
Continual Learning enables models to learn and adapt to new tasks while retaining prior knowledge. Introducing new tasks, however, can naturally lead to feature entanglement across tasks, limiting the model's capability to distinguish between new domain data. In this work, we propose a method called Feature Realignment through Experts on hyperSpHere in Continual Learning (Fresh-CL). By leveraging predefined and fixed simplex equiangular tight frame (ETF) classifiers on a hypersphere, our model improves feature separation both intra and inter tasks. However, the projection to a simplex ETF shifts with new tasks, disrupting structured feature representation of previous tasks and degrading performance. Therefore, we propose a dynamic extension of ETF through mixture of experts, enabling adaptive projections onto diverse subspaces to enhance feature representation. Experiments on 11 datasets demonstrate a 2% improvement in accuracy compared to the strongest baseline, particularly in fine-grained datasets, confirming the efficacy of combining ETF and MoE to improve feature distinction in continual learning scenarios.
comment: Accepted by ICASSP 2025
♻ ☆ Mitigating Low-Frequency Bias: Feature Recalibration and Frequency Attention Regularization for Adversarial Robustness
Ensuring the robustness of deep neural networks against adversarial attacks remains a fundamental challenge in computer vision. While adversarial training (AT) has emerged as a promising defense strategy, our analysis reveals a critical limitation: AT-trained models exhibit a bias toward low-frequency features while neglecting high-frequency components. This bias is particularly concerning as each frequency component carries distinct and crucial information: low-frequency features encode fundamental structural patterns, while high-frequency features capture intricate details and textures. To address this limitation, we propose High-Frequency Feature Disentanglement and Recalibration (HFDR), a novel module that strategically separates and recalibrates frequency-specific features to capture latent semantic cues. We further introduce frequency attention regularization to harmonize feature extraction across the frequency spectrum and mitigate the inherent low-frequency bias of AT. Extensive experiments demonstrate our method's superior performance against white-box attacks and transfer attacks, while exhibiting strong generalization capabilities across diverse scenarios.
♻ ☆ SELMA3D challenge: Self-supervised learning for 3D light-sheet microscopy image segmentation
Recent innovations in light sheet microscopy, paired with developments in tissue clearing techniques, enable the 3D imaging of large mammalian tissues with cellular resolution. Combined with the progress in large-scale data analysis, driven by deep learning, these innovations empower researchers to rapidly investigate the morphological and functional properties of diverse biological samples. Segmentation, a crucial preliminary step in the analysis process, can be automated using domain-specific deep learning models with expert-level performance. However, these models exhibit high sensitivity to domain shifts, leading to a significant drop in accuracy when applied to data outside their training distribution. To address this limitation, and inspired by the recent success of self-supervised learning in training generalizable models, we organized the SELMA3D Challenge during the MICCAI 2024 conference. SELMA3D provides a vast collection of light-sheet images from cleared mice and human brains, comprising 35 large 3D images-each with over 1000^3 voxels-and 315 annotated small patches for finetuning, preliminary testing and final testing. The dataset encompasses diverse biological structures, including vessel-like and spot-like structures. Five teams participated in all phases of the challenge, and their proposed methods are reviewed in this paper. Quantitative and qualitative results from most participating teams demonstrate that self-supervised learning on large datasets improves segmentation model performance and generalization. We will continue to support and extend SELMA3D as an inaugural MICCAI challenge focused on self-supervised learning for 3D microscopy image segmentation.
comment: 2st version
Computation and Language 42
☆ Harnessing Large Language Models for Disaster Management: A Survey
Large language models (LLMs) have revolutionized scientific research with their exceptional capabilities and transformed various fields. Among their practical applications, LLMs have been playing a crucial role in mitigating threats to human life, infrastructure, and the environment. Despite growing research in disaster LLMs, there remains a lack of systematic review and in-depth analysis of LLMs for natural disaster management. To address the gap, this paper presents a comprehensive survey of existing LLMs in natural disaster management, along with a taxonomy that categorizes existing works based on disaster phases and application scenarios. By collecting public datasets and identifying key challenges and opportunities, this study aims to guide the professional community in developing advanced LLMs for disaster management to enhance the resilience against natural disasters.
☆ Risk-Averse Finetuning of Large Language Models
We consider the challenge of mitigating the generation of negative or toxic content by the Large Language Models (LLMs) in response to certain prompts. We propose integrating risk-averse principles into LLM fine-tuning to minimize the occurrence of harmful outputs, particularly rare but significant events. By optimizing the risk measure of Conditional Value at Risk (CVaR), our methodology trains LLMs to exhibit superior performance in avoiding toxic outputs while maintaining effectiveness in generative tasks. Empirical evaluations on sentiment modification and toxicity mitigation tasks demonstrate the efficacy of risk-averse reinforcement learning with human feedback (RLHF) in promoting a safer and more constructive online discourse environment.
comment: Neurips 2024
☆ Language Fusion for Parameter-Efficient Cross-lingual Transfer
Limited availability of multilingual text corpora for training language models often leads to poor performance on downstream tasks due to undertrained representation spaces for languages other than English. This 'under-representation' has motivated recent cross-lingual transfer methods to leverage the English representation space by e.g. mixing English and 'non-English' tokens at the input level or extending model parameters to accommodate new languages. However, these approaches often come at the cost of increased computational complexity. We propose Fusion forLanguage Representations (FLARE) in adapters, a novel method that enhances representation quality and downstream performance for languages other than English while maintaining parameter efficiency. FLARE integrates source and target language representations within low-rank (LoRA) adapters using lightweight linear transformations, maintaining parameter efficiency while improving transfer performance. A series of experiments across representative cross-lingual natural language understanding tasks, including natural language inference, question-answering and sentiment analysis, demonstrate FLARE's effectiveness. FLARE achieves performance improvements of 4.9% for Llama 3.1 and 2.2% for Gemma~2 compared to standard LoRA fine-tuning on question-answering tasks, as measured by the exact match metric.
comment: 20 pages
☆ Causal Claims in Economics
We analyze over 44,000 NBER and CEPR working papers from 1980 to 2023 using a custom language model to construct knowledge graphs that map economic concepts and their relationships. We distinguish between general claims and those documented via causal inference methods (e.g., DiD, IV, RDD, RCTs). We document a substantial rise in the share of causal claims-from roughly 4% in 1990 to nearly 28% in 2020-reflecting the growing influence of the "credibility revolution." We find that causal narrative complexity (e.g., the depth of causal chains) strongly predicts both publication in top-5 journals and higher citation counts, whereas non-causal complexity tends to be uncorrelated or negatively associated with these outcomes. Novelty is also pivotal for top-5 publication, but only when grounded in credible causal methods: introducing genuinely new causal edges or paths markedly increases both the likelihood of acceptance at leading outlets and long-run citations, while non-causal novelty exhibits weak or even negative effects. Papers engaging with central, widely recognized concepts tend to attract more citations, highlighting a divergence between factors driving publication success and long-term academic impact. Finally, bridging underexplored concept pairs is rewarded primarily when grounded in causal methods, yet such gap filling exhibits no consistent link with future citations. Overall, our findings suggest that methodological rigor and causal innovation are key drivers of academic recognition, but sustained impact may require balancing novel contributions with conceptual integration into established economic discourse.
comment: For data, interactive tools, and additional project information, visit https://www.causal.claims/. The website contains resources such as data downloads, interactive author and paper-level knowledge graphs, and more
☆ Transfer Learning of Tabular Data by Finetuning Large Language Models
Despite the artificial intelligence (AI) revolution, deep learning has yet to achieve much success with tabular data due to heterogeneous feature space and limited sample sizes without viable transfer learning. The new era of generative AI, powered by large language models (LLM), brings unprecedented learning opportunities to diverse data and domains. This paper investigates the effectiveness of an LLM application programming interface (API) and transfer learning of LLM in tabular data classification. LLM APIs respond to input text prompts with tokenized data and instructions, whereas transfer learning finetunes an LLM for a target classification task. This paper proposes an end-to-end finetuning of LLM to demonstrate cross-data transfer learning on ten benchmark data sets when large pre-trained tabular data models do not exist to facilitate transfer learning. The proposed LLM finetuning method outperforms state-of-the-art machine and deep learning methods on tabular data with less than ten features - a standard feature size for tabular data sets. The transfer learning approach uses a fraction of the computational cost of other deep learning or API-based solutions while ensuring competitive or superior classification performance.
☆ A Comprehensive Evaluation of Large Language Models on Mental Illnesses in Arabic Context
Mental health disorders pose a growing public health concern in the Arab world, emphasizing the need for accessible diagnostic and intervention tools. Large language models (LLMs) offer a promising approach, but their application in Arabic contexts faces challenges including limited labeled datasets, linguistic complexity, and translation biases. This study comprehensively evaluates 8 LLMs, including general multi-lingual models, as well as bi-lingual ones, on diverse mental health datasets (such as AraDepSu, Dreaddit, MedMCQA), investigating the impact of prompt design, language configuration (native Arabic vs. translated English, and vice versa), and few-shot prompting on diagnostic performance. We find that prompt engineering significantly influences LLM scores mainly due to reduced instruction following, with our structured prompt outperforming a less structured variant on multi-class datasets, with an average difference of 14.5\%. While language influence on performance was modest, model selection proved crucial: Phi-3.5 MoE excelled in balanced accuracy, particularly for binary classification, while Mistral NeMo showed superior performance in mean absolute error for severity prediction tasks. Few-shot prompting consistently improved performance, with particularly substantial gains observed for GPT-4o Mini on multi-class classification, boosting accuracy by an average factor of 1.58. These findings underscore the importance of prompt optimization, multilingual analysis, and few-shot learning for developing culturally sensitive and effective LLM-based mental health tools for Arabic-speaking populations.
☆ A General Framework for Inference-time Scaling and Steering of Diffusion Models
Diffusion models produce impressive results in modalities ranging from images and video to protein design and text. However, generating samples with user-specified properties remains a challenge. Recent research proposes fine-tuning models to maximize rewards that capture desired properties, but these methods require expensive training and are prone to mode collapse. In this work, we propose Feynman Kac (FK) steering, an inference-time framework for steering diffusion models with reward functions. FK steering works by sampling a system of multiple interacting diffusion processes, called particles, and resampling particles at intermediate steps based on scores computed using functions called potentials. Potentials are defined using rewards for intermediate states and are selected such that a high value indicates that the particle will yield a high-reward sample. We explore various choices of potentials, intermediate rewards, and samplers. We evaluate FK steering on text-to-image and text diffusion models. For steering text-to-image models with a human preference reward, we find that FK steering a 0.8B parameter model outperforms a 2.6B parameter fine-tuned model on prompt fidelity, with faster sampling and no training. For steering text diffusion models with rewards for text quality and specific text attributes, we find that FK steering generates lower perplexity, more linguistically acceptable outputs and enables gradient-free control of attributes like toxicity. Our results demonstrate that inference-time scaling and steering of diffusion models, even with off-the-shelf rewards, can provide significant sample quality gains and controllability benefits. Code is available at https://github.com/zacharyhorvitz/Fk-Diffusion-Steering .
☆ SPAM: Spike-Aware Adam with Momentum Reset for Stable LLM Training
Large Language Models (LLMs) have demonstrated exceptional performance across diverse tasks, yet their training remains highly resource-intensive and susceptible to critical challenges such as training instability. A predominant source of this instability stems from gradient and loss spikes, which disrupt the learning process, often leading to costly interventions like checkpoint recovery and experiment restarts, further amplifying inefficiencies. This paper presents a comprehensive investigation into gradient spikes observed during LLM training, revealing their prevalence across multiple architectures and datasets. Our analysis shows that these spikes can be up to $1000\times$ larger than typical gradients, substantially deteriorating model performance. To address this issue, we propose Spike-Aware Adam with Momentum Reset SPAM, a novel optimizer designed to counteract gradient spikes through momentum reset and spike-aware gradient clipping. Extensive experiments, including both pre-training and fine-tuning, demonstrate that SPAM consistently surpasses Adam and its variants across various tasks, including (1) LLM pre-training from 60M to 1B, (2) 4-bit LLM pre-training,(3) reinforcement learning, and (4) Time Series Forecasting. Additionally, SPAM facilitates memory-efficient training by enabling sparse momentum, where only a subset of momentum terms are maintained and updated. When operating under memory constraints, SPAM outperforms state-of-the-art memory-efficient optimizers such as GaLore and Adam-Mini. Our work underscores the importance of mitigating gradient spikes in LLM training and introduces an effective optimization strategy that enhances both training stability and resource efficiency at scale. Code is available at https://github.com/TianjinYellow/SPAM-Optimizer.git
☆ LLMs Model Non-WEIRD Populations: Experiments with Synthetic Cultural Agents
Despite its importance, studying economic behavior across diverse, non-WEIRD (Western, Educated, Industrialized, Rich, and Democratic) populations presents significant challenges. We address this issue by introducing a novel methodology that uses Large Language Models (LLMs) to create synthetic cultural agents (SCAs) representing these populations. We subject these SCAs to classic behavioral experiments, including the dictator and ultimatum games. Our results demonstrate substantial cross-cultural variability in experimental behavior. Notably, for populations with available data, SCAs' behaviors qualitatively resemble those of real human subjects. For unstudied populations, our method can generate novel, testable hypotheses about economic behavior. By integrating AI into experimental economics, this approach offers an effective and ethical method to pilot experiments and refine protocols for hard-to-reach populations. Our study provides a new tool for cross-cultural economic studies and demonstrates how LLMs can help experimental behavioral research.
☆ Correcting Annotator Bias in Training Data: Population-Aligned Instance Replication (PAIR)
Models trained on crowdsourced labels may not reflect broader population views when annotator pools are not representative. Since collecting representative labels is challenging, we propose Population-Aligned Instance Replication (PAIR), a method to address this bias through statistical adjustment. Using a simulation study of hate speech and offensive language detection, we create two types of annotators with different labeling tendencies and generate datasets with varying proportions of the types. Models trained on unbalanced annotator pools show poor calibration compared to those trained on representative data. However, PAIR, which duplicates labels from underrepresented annotator groups to match population proportions, significantly reduces bias without requiring new data collection. These results suggest statistical techniques from survey research can help align model training with target populations even when representative annotator pools are unavailable. We conclude with three practical recommendations for improving training data quality.
☆ Event Argument Extraction with Enriched Prompts
This work aims to delve deeper into prompt-based event argument extraction (EAE) models. We explore the impact of incorporating various types of information into the prompt on model performance, including trigger, other role arguments for the same event, and role arguments across multiple events within the same document. Further, we provide the best possible performance that the prompt-based EAE model can attain and demonstrate such models can be further optimized from the perspective of the training objective. Experiments are carried out on three small language models and two large language models in RAMS.
☆ Improving Cross-Lingual Phonetic Representation of Low-Resource Languages Through Language Similarity Analysis ICASSP 2025
This paper examines how linguistic similarity affects cross-lingual phonetic representation in speech processing for low-resource languages, emphasizing effective source language selection. Previous cross-lingual research has used various source languages to enhance performance for the target low-resource language without thorough consideration of selection. Our study stands out by providing an in-depth analysis of language selection, supported by a practical approach to assess phonetic proximity among multiple language families. We investigate how within-family similarity impacts performance in multilingual training, which aids in understanding language dynamics. We also evaluate the effect of using phonologically similar languages, regardless of family. For the phoneme recognition task, utilizing phonologically similar languages consistently achieves a relative improvement of 55.6% over monolingual training, even surpassing the performance of a large-scale self-supervised learning model. Multilingual training within the same language family demonstrates that higher phonological similarity enhances performance, while lower similarity results in degraded performance compared to monolingual training.
comment: 10 pages, 5 figures, accepted to ICASSP 2025
☆ Bridging the Fairness Gap: Enhancing Pre-trained Models with LLM-Generated Sentences
Pre-trained language models (PLMs) are trained on data that inherently contains gender biases, leading to undesirable impacts. Traditional debiasing methods often rely on external corpora, which may lack quality, diversity, or demographic balance, affecting the effectiveness of debiasing. With the rise of large language models and their extensive knowledge, we propose enhancing fairness (Fair-Gender) in PLMs by absorbing coherent, attribute-balanced, and semantically rich sentences. However, these sentences cannot be directly used for debiasing due to alignment issues and the risk of negative transfer. We address this by applying causal analysis to estimate causal effects, filtering out unaligned sentences, and identifying aligned ones for incorporation into PLMs, thereby ensuring positive transfer. Experiments show that our approach significantly reduces gender biases in PLMs while preserving their language expressiveness.
☆ 3DCoMPaT200: Language-Grounded Compositional Understanding of Parts and Materials of 3D Shapes
Understanding objects in 3D at the part level is essential for humans and robots to navigate and interact with the environment. Current datasets for part-level 3D object understanding encompass a limited range of categories. For instance, the ShapeNet-Part and PartNet datasets only include 16, and 24 object categories respectively. The 3DCoMPaT dataset, specifically designed for compositional understanding of parts and materials, contains only 42 object categories. To foster richer and fine-grained part-level 3D understanding, we introduce 3DCoMPaT200, a large-scale dataset tailored for compositional understanding of object parts and materials, with 200 object categories with $\approx$5 times larger object vocabulary compared to 3DCoMPaT and $\approx$ 4 times larger part categories. Concretely, 3DCoMPaT200 significantly expands upon 3DCoMPaT, featuring 1,031 fine-grained part categories and 293 distinct material classes for compositional application to 3D object parts. Additionally, to address the complexities of compositional 3D modeling, we propose a novel task of Compositional Part Shape Retrieval using ULIP to provide a strong 3D foundational model for 3D Compositional Understanding. This method evaluates the model shape retrieval performance given one, three, or six parts described in text format. These results show that the model's performance improves with an increasing number of style compositions, highlighting the critical role of the compositional dataset. Such results underscore the dataset's effectiveness in enhancing models' capability to understand complex 3D shapes from a compositional perspective. Code and Data can be found at http://github.com/3DCoMPaT200/3DCoMPaT200
☆ Padding Tone: A Mechanistic Analysis of Padding Tokens in T2I Models
Text-to-image (T2I) diffusion models rely on encoded prompts to guide the image generation process. Typically, these prompts are extended to a fixed length by adding padding tokens before text encoding. Despite being a default practice, the influence of padding tokens on the image generation process has not been investigated. In this work, we conduct the first in-depth analysis of the role padding tokens play in T2I models. We develop two causal techniques to analyze how information is encoded in the representation of tokens across different components of the T2I pipeline. Using these techniques, we investigate when and how padding tokens impact the image generation process. Our findings reveal three distinct scenarios: padding tokens may affect the model's output during text encoding, during the diffusion process, or be effectively ignored. Moreover, we identify key relationships between these scenarios and the model's architecture (cross or self-attention) and its training process (frozen or trained text encoder). These insights contribute to a deeper understanding of the mechanisms of padding tokens, potentially informing future model design and training practices in T2I systems.
☆ Hierarchical Divide-and-Conquer for Fine-Grained Alignment in LLM-Based Medical Evaluation
In the rapidly evolving landscape of large language models (LLMs) for medical applications, ensuring the reliability and accuracy of these models in clinical settings is paramount. Existing benchmarks often focus on fixed-format tasks like multiple-choice QA, which fail to capture the complexity of real-world clinical diagnostics. Moreover, traditional evaluation metrics and LLM-based evaluators struggle with misalignment, often providing oversimplified assessments that do not adequately reflect human judgment. To address these challenges, we introduce HDCEval, a Hierarchical Divide-and-Conquer Evaluation framework tailored for fine-grained alignment in medical evaluation. HDCEval is built on a set of fine-grained medical evaluation guidelines developed in collaboration with professional doctors, encompassing Patient Question Relevance, Medical Knowledge Correctness, and Expression. The framework decomposes complex evaluation tasks into specialized subtasks, each evaluated by expert models trained through Attribute-Driven Token Optimization (ADTO) on a meticulously curated preference dataset. This hierarchical approach ensures that each aspect of the evaluation is handled with expert precision, leading to a significant improvement in alignment with human evaluators.
☆ ZOQO: Zero-Order Quantized Optimization ICASSP 2025
The increasing computational and memory demands in deep learning present significant challenges, especially in resource-constrained environments. We introduce a zero-order quantized optimization (ZOQO) method designed for training models with quantized parameters and operations. Our approach leverages zero-order approximations of the gradient sign and adapts the learning process to maintain the parameters' quantization without the need for full-precision gradient calculations. We demonstrate the effectiveness of ZOQO through experiments in fine-tuning of large language models and black-box adversarial attacks. Despite the limitations of zero-order and quantized operations training, our method achieves competitive performance compared to full-precision methods, highlighting its potential for low-resource environments.
comment: Accepted to ICASSP 2025
☆ Better Prompt Compression Without Multi-Layer Perceptrons
Prompt compression is a promising approach to speeding up language model inference without altering the generative model. Prior works compress prompts into smaller sequences of learned tokens using an encoder that is trained as a LowRank Adaptation (LoRA) of the inference language model. However, we show that the encoder does not need to keep the original language model's architecture to achieve useful compression. We introduce the Attention-Only Compressor (AOC), which learns a prompt compression encoder after removing the multilayer perceptron (MLP) layers in the Transformer blocks of a language model, resulting in an encoder with roughly 67% less parameters compared to the original model. Intriguingly we find that, across a range of compression ratios up to 480x, AOC can better regenerate prompts and outperform a baseline compression encoder that is a LoRA of the inference language model without removing MLP layers. These results demonstrate that the architecture of prompt compression encoders does not need to be identical to that of the original decoder language model, paving the way for further research into architectures and approaches for prompt compression.
comment: 7 pages, 0 figures
☆ Measuring the Robustness of Reference-Free Dialogue Evaluation Systems
Advancements in dialogue systems powered by large language models (LLMs) have outpaced the development of reliable evaluation metrics, particularly for diverse and creative responses. We present a benchmark for evaluating the robustness of reference-free dialogue metrics against four categories of adversarial attacks: speaker tag prefixes, static responses, ungrammatical responses, and repeated conversational context. We analyze metrics such as DialogRPT, UniEval, and PromptEval -- a prompt-based method leveraging LLMs -- across grounded and ungrounded datasets. By examining both their correlation with human judgment and susceptibility to adversarial attacks, we find that these two axes are not always aligned; metrics that appear to be equivalent when judged by traditional benchmarks may, in fact, vary in their scores of adversarial responses. These findings motivate the development of nuanced evaluation frameworks to address real-world dialogue challenges.
☆ ZNO-Eval: Benchmarking reasoning capabilities of large language models in Ukrainian
As the usage of large language models for problems outside of simple text understanding or generation increases, assessing their abilities and limitations becomes crucial. While significant progress has been made in this area over the last few years, most research has focused on benchmarking English, leaving other languages underexplored. This makes evaluating the reasoning and robustness level of language models in Ukrainian particularly challenging. The purpose of this work is to establish a comprehensive benchmark for the reasoning capabilities evaluation of large language models in the Ukrainian language. This paper presents the ZNO-Eval benchmark based on real exam tasks from Ukraine's standardized educational testing system: the External Independent Evaluation and the National Multi-subject Test. With single-answer options, multiple-choice, matching, and open-ended questions from diverse subjects, including Ukrainian language, mathematics, history, and geography, this dataset paves the way toward a thorough analysis of reasoning capabilities across different domains and complexities. Evaluation of several well-known language models, such as GPT-3.5-Turbo, GPT-4o, GPT-4-Turbo, Mistral Large, Claude 3 Opus, and Gemini-1.5 Pro on this benchmark demonstrated the superiority of GPT-4o in both common knowledge reasoning and intricate language tasks. At the same time, Gemini Pro and GPT-4 Turbo excelled in the arithmetic domain, leading in single-answer and open-ended math problems. While all models were close to max performance in text-only common knowledge tasks like history and geography, there still is a gap for Ukrainian language and math, thus highlighting the importance of developing specialized language benchmarks for more accurate assessments of model capabilities and limitations across different languages and contexts.
comment: 7 pages, 5 figures. X International conference "Informatics. Culture. Technology." (2024)
☆ Fine-tuning ChatGPT for Automatic Scoring of Written Scientific Explanations in Chinese
The development of explanations for scientific phenomena is essential in science assessment, but scoring student-written explanations remains challenging and resource-intensive. Large language models (LLMs) have shown promise in addressing this issue, particularly in alphabetic languages like English. However, their applicability to logographic languages is less explored. This study investigates the potential of fine-tuning ChatGPT, a leading LLM, to automatically score scientific explanations written in Chinese. Student responses to seven scientific explanation tasks were collected and automatically scored, with scoring accuracy examined in relation to reasoning complexity using the Kendall correlation. A qualitative analysis explored how linguistic features influenced scoring accuracy. The results show that domain-specific adaptation enables ChatGPT to score Chinese scientific explanations with accuracy. However, scoring accuracy correlates with reasoning complexity: a negative correlation for lower-level responses and a positive one for higher-level responses. The model overrates complex reasoning in low-level responses with intricate sentence structures and underrates high-level responses using concise causal reasoning. These correlations stem from linguistic features--simplicity and clarity enhance accuracy for lower-level responses, while comprehensiveness improves accuracy for higher-level ones. Simpler, shorter responses tend to score more accurately at lower levels, whereas longer, information-rich responses yield better accuracy at higher levels. These findings demonstrate the effectiveness of LLMs in automatic scoring within a Chinese context and emphasize the importance of linguistic features and reasoning complexity in fine-tuning scoring models for educational assessments.
☆ TAPO: Task-Referenced Adaptation for Prompt Optimization ICASSP 2025
Prompt engineering can significantly improve the performance of large language models (LLMs), with automated prompt optimization (APO) gaining significant attention due to the time-consuming and laborious nature of manual prompt design. However, much of the existing work in APO overlooks task-specific characteristics, resulting in prompts that lack domain specificity and are not well-suited for task-specific optimization. In this paper, we introduce TAPO, a multitask-aware prompt optimization framework composed of three key modules. First, a task-aware metric selection module is proposed to enhance task-specific prompt generation capabilities. Second, we present a multi-metrics evaluation module to jointly evaluate prompts from multiple perspectives. Third, an evolution-based optimization framework is introduced for automatic prompt refinement, which improves adaptability across various tasks. Extensive experiments on six datasets demonstrate the effectiveness of our approach, and our code is publicly available.
comment: Accepted to ICASSP 2025
♻ ☆ ChatGPT Needs SPADE (Sustainability, PrivAcy, Digital divide, and Ethics) Evaluation: A Review
ChatGPT is another large language model (LLM) vastly available for the consumers on their devices but due to its performance and ability to converse effectively, it has gained a huge popularity amongst research as well as industrial community. Recently, many studies have been published to show the effectiveness, efficiency, integration, and sentiments of chatGPT and other LLMs. In contrast, this study focuses on the important aspects that are mostly overlooked, i.e. sustainability, privacy, digital divide, and ethics and suggests that not only chatGPT but every subsequent entry in the category of conversational bots should undergo Sustainability, PrivAcy, Digital divide, and Ethics (SPADE) evaluation. This paper discusses in detail the issues and concerns raised over chatGPT in line with aforementioned characteristics. We also discuss the recent EU AI Act briefly in accordance with the SPADE evaluation. We support our hypothesis by some preliminary data collection and visualizations along with hypothesized facts. We also suggest mitigations and recommendations for each of the concerns. Furthermore, we also suggest some policies and recommendations for EU AI policy act concerning ethics, digital divide, and sustainability
comment: 29 pages, 8 figures, 4 tables
♻ ☆ Explaining Datasets in Words: Statistical Models with Natural Language Parameters
To make sense of massive data, we often fit simplified models and then interpret the parameters; for example, we cluster the text embeddings and then interpret the mean parameters of each cluster. However, these parameters are often high-dimensional and hard to interpret. To make model parameters directly interpretable, we introduce a family of statistical models -- including clustering, time series, and classification models -- parameterized by natural language predicates. For example, a cluster of text about COVID could be parameterized by the predicate "discusses COVID". To learn these statistical models effectively, we develop a model-agnostic algorithm that optimizes continuous relaxations of predicate parameters with gradient descent and discretizes them by prompting language models (LMs). Finally, we apply our framework to a wide range of problems: taxonomizing user chat dialogues, characterizing how they evolve across time, finding categories where one language model is better than the other, clustering math problems based on subareas, and explaining visual features in memorable images. Our framework is highly versatile, applicable to both textual and visual domains, can be easily steered to focus on specific properties (e.g. subareas), and explains sophisticated concepts that classical methods (e.g. n-gram analysis) struggle to produce.
♻ ☆ TrustRAG: Enhancing Robustness and Trustworthiness in RAG
Retrieval-Augmented Generation (RAG) systems enhance large language models (LLMs) by integrating external knowledge sources, enabling more accurate and contextually relevant responses tailored to user queries. However, these systems remain vulnerable to corpus poisoning attacks that can significantly degrade LLM performance through the injection of malicious content. To address these challenges, we propose TrustRAG, a robust framework that systematically filters compromised and irrelevant contents before they are retrieved for generation. Our approach implements a two-stage defense mechanism: At the first stage, it employs K-means clustering to identify potential attack patterns in retrieved documents using cosine similarity and ROUGE metrics as guidance, effectively isolating suspicious content. Secondly, it performs a self-assessment which detects malicious documents and resolves discrepancies between the model's internal knowledge and external information. TrustRAG functions as a plug-and-play, training-free module that integrates seamlessly with any language model, whether open or closed-source. In addition, TrustRAG maintains high contextual relevance while strengthening defenses against corpus poisoning attacks. Through extensive experimental validation, we demonstrate that TrustRAG delivers substantial improvements in retrieval accuracy, efficiency, and attack resistance compared to existing approaches across multiple model architectures and datasets. We have made TrustRAG available as open-source software at \url{https://github.com/HuichiZhou/TrustRAG}.
♻ ☆ Unlocking the Power of LLM Uncertainty for Active In-Context Example Selection
Large Language Models (LLMs) have shown remarkable performance across a wide range of downstream tasks. However, it is challenging for users to discern whether the responses of LLM are generated with certainty or are fabricated to meet user expectations. In this paper, we introduce Uncertainty Tripartite Testing Paradigm (Unc-TTP), a novel method for classifying LLM uncertainty by leveraging output inconsistency. Specifically, Unc-TTP performs three rounds of sampling under varying label injection interference, enumerating all possible outcomes, and uses the degree of output inconsistency as the indicator of the LLM's intrinsic uncertainty. To validate the effectiveness of this inconsistency-defined uncertainty, we draw inspiration from Active Learning, comparing the informativeness of actively selected in-context examples. Our experiments show that uncertainty examples selected via Unc-TTP are more informative than certainty examples. Furthermore, the Unc-TTP-guided uncertainty-based active example selection strategy outperforms existing methods, highlighting its effectiveness in classifying LLM uncertainty and enhancing in-context learning. This work not only underscores the potential of inconsistency-based uncertainty classification for both open- and closed-source LLMs but also presents a practical approach for leveraging uncertainty to improve LLM performance in real-world tasks.
♻ ☆ ObfuscaTune: Obfuscated Offsite Fine-tuning and Inference of Proprietary LLMs on Private Datasets AAAI 2025
This work addresses the timely yet underexplored problem of performing inference and finetuning of a proprietary LLM owned by a model provider entity on the confidential/private data of another data owner entity, in a way that ensures the confidentiality of both the model and the data. Hereby, the finetuning is conducted offsite, i.e., on the computation infrastructure of a third-party cloud provider. We tackle this problem by proposing ObfuscaTune, a novel, efficient and fully utility-preserving approach that combines a simple yet effective obfuscation technique with an efficient usage of confidential computing (only 5% of the model parameters are placed on TEE). We empirically demonstrate the effectiveness of ObfuscaTune by validating it on GPT-2 models with different sizes on four NLP benchmark datasets. Finally, we compare to a na\"ive version of our approach to highlight the necessity of using random matrices with low condition numbers in our approach to reduce errors induced by the obfuscation.
comment: Accepted at AAAI 2025 (PPAI Workshop)
♻ ☆ What Languages are Easy to Language-Model? A Perspective from Learning Probabilistic Regular Languages ACL 2024
What can large language models learn? By definition, language models (LM) are distributions over strings. Therefore, an intuitive way of addressing the above question is to formalize it as a matter of learnability of classes of distributions over strings. While prior work in this direction focused on assessing the theoretical limits, in contrast, we seek to understand the empirical learnability. Unlike prior empirical work, we evaluate neural LMs on their home turf-learning probabilistic languages-rather than as classifiers of formal languages. In particular, we investigate the learnability of regular LMs (RLMs) by RNN and Transformer LMs. We empirically test the learnability of RLMs as a function of various complexity parameters of the RLM and the hidden state size of the neural LM. We find that the RLM rank, which corresponds to the size of linear space spanned by the logits of its conditional distributions, and the expected length of sampled strings are strong and significant predictors of learnability for both RNNs and Transformers. Several other predictors also reach significance, but with differing patterns between RNNs and Transformers.
comment: Accepted to ACL 2024
♻ ☆ Semantic Prompt Learning for Weakly-Supervised Semantic Segmentation WACV 2025
Weakly-Supervised Semantic Segmentation (WSSS) aims to train segmentation models using image data with only image-level supervision. Since precise pixel-level annotations are not accessible, existing methods typically focus on producing pseudo masks for training segmentation models by refining CAM-like heatmaps. However, the produced heatmaps may capture only the discriminative image regions of object categories or the associated co-occurring backgrounds. To address the issues, we propose a Semantic Prompt Learning for WSSS (SemPLeS) framework, which learns to effectively prompt the CLIP latent space to enhance the semantic alignment between the segmented regions and the target object categories. More specifically, we propose Contrastive Prompt Learning and Prompt-guided Semantic Refinement to learn the prompts that adequately describe and suppress the co-occurring backgrounds associated with each object category. In this way, SemPLeS can perform better semantic alignment between object regions and class labels, resulting in desired pseudo masks for training segmentation models. The proposed SemPLeS framework achieves competitive performance on standard WSSS benchmarks, PASCAL VOC 2012 and MS COCO 2014, and shows compatibility with other WSSS methods. Code: https://github.com/NVlabs/SemPLeS.
comment: WACV 2025. Code: https://github.com/NVlabs/SemPLeS. Project page: https://projectdisr.github.io/semples/
♻ ☆ Accurate and Regret-aware Numerical Problem Solver for Tabular Question Answering
Question answering on free-form tables (a.k.a. TableQA) is a challenging task because of the flexible structure and complex schema of tables. Recent studies use Large Language Models (LLMs) for this task, exploiting their capability in understanding the questions and tabular data, which are typically given in natural language and contain many textual fields, respectively. While this approach has shown promising results, it overlooks the challenges brought by numerical values which are common in tabular data, and LLMs are known to struggle with such values. We aim to address this issue, and we propose a model named TabLaP that uses LLMs as a planner rather than an answer generator. This approach exploits LLMs' capability in multi-step reasoning while leaving the actual numerical calculations to a Python interpreter for accurate calculation. Recognizing the inaccurate nature of LLMs, we further make a first attempt to quantify the trustworthiness of the answers produced by TabLaP, such that users can use TabLaP in a regret-aware manner. Experimental results on two benchmark datasets show that TabLaP is substantially more accurate than the state-of-the-art models, improving the answer accuracy by 5.7% and 5.8% on the two datasets, respectively.
♻ ☆ Improving Zero-Shot Chinese-English Code-Switching ASR with kNN-CTC and Gated Monolingual Datastores ICASSP 2025
The kNN-CTC model has proven to be effective for monolingual automatic speech recognition (ASR). However, its direct application to multilingual scenarios like code-switching, presents challenges. Although there is potential for performance improvement, a kNN-CTC model utilizing a single bilingual datastore can inadvertently introduce undesirable noise from the alternative language. To address this, we propose a novel kNN-CTC-based code-switching ASR (CS-ASR) framework that employs dual monolingual datastores and a gated datastore selection mechanism to reduce noise interference. Our method selects the appropriate datastore for decoding each frame, ensuring the injection of language-specific information into the ASR process. We apply this framework to cutting-edge CTC-based models, developing an advanced CS-ASR system. Extensive experiments demonstrate the remarkable effectiveness of our gated datastore mechanism in enhancing the performance of zero-shot Chinese-English CS-ASR.
comment: Accepted by ICASSP 2025
♻ ☆ Unsupervised Word Discovery: Boundary Detection with Clustering vs. Dynamic Programming ICASSP 2025
We look at the long-standing problem of segmenting unlabeled speech into word-like segments and clustering these into a lexicon. Several previous methods use a scoring model coupled with dynamic programming to find an optimal segmentation. Here we propose a much simpler strategy: we predict word boundaries using the dissimilarity between adjacent self-supervised features, then we cluster the predicted segments to construct a lexicon. For a fair comparison, we update the older ES-KMeans dynamic programming method with better features and boundary constraints. On the five-language ZeroSpeech benchmarks, our simple approach gives similar state-of-the-art results compared to the new ES-KMeans+ method, while being almost five times faster. Project webpage: https://s-malan.github.io/prom-seg-clus.
comment: Accepted at ICASSP 2025
♻ ☆ Linear Attention Sequence Parallelism
Sequence parallelism (SP) serves as a prevalent strategy to handle long sequences that exceed the memory limit of a single device. However, for linear sequence modeling methods like linear attention, existing SP approaches do not take advantage of their right-product-first feature, resulting in sub-optimal communication efficiency and usability. In this paper, we introduce Linear Attention Sequence Parallelism (LASP), an efficient SP approach designed for linear attention-based transformer models. Specifically, we design an efficient point-to-point ring-style communication mechanism to leverage the right-product kernel trick of linear attention, which sharply decreases the communication overhead, comparing with existing SP methods. We enhance the computation efficiency of LASP by performing kernel fusion and intermediate state caching, making the implementation of LASP hardware-friendly on GPUs. Furthermore, we meticulously ensure the compatibility of sequence-level LASP with all types of batch-level data parallel methods, which is vital for distributed training on large clusters with very-long sequences. We also discuss the generalization of LASP on other linear sequence modeling methods. Extensive experiments on linear attention-based models are conducted with varying sequence lengths from 2K to 4096K. LASP scales sequence length up to 4096K on 128 GPUs, which is 8$\times$ longer than existing SP methods. The code is available at https://github.com/OpenNLPLab/LASP.
comment: Technical report, 20 pages
♻ ☆ What Are They Doing? Joint Audio-Speech Co-Reasoning ICASSP 2025
In audio and speech processing, tasks usually focus on either the audio or speech modality, even when both sounds and human speech are present in the same audio clip. Recent Auditory Large Language Models (ALLMs) have made it possible to process audio and speech simultaneously within a single model, leading to further considerations of joint audio-speech tasks. In this paper, we establish a novel benchmark to investigate how well ALLMs can perform joint audio-speech processing. Specifically, we introduce Joint Audio-Speech Co-Reasoning (JASCO), a novel task that unifies audio and speech processing, strictly requiring co-reasoning across both modalities. We also release a scene-reasoning dataset called "What Are They Doing". Additionally, we provide deeper insights into the models' behaviors by analyzing their dependence on each modality.
comment: Accepted to ICASSP 2025
♻ ☆ Towards Understanding Multi-Task Learning (Generalization) of LLMs via Detecting and Exploring Task-Specific Neurons
While large language models (LLMs) have demonstrated superior multi-task capabilities, understanding the learning mechanisms behind this is still a challenging problem. In this paper, we attempt to understand such mechanisms from the perspective of neurons. Specifically, we detect task-sensitive neurons in LLMs via gradient attribution on task-specific data. Through extensive deactivation and fine-tuning experiments, we demonstrate that the detected neurons are highly correlated with the given task, which we term as task-specific neurons. With these identified task-specific neurons, we delve into two common problems in multi-task learning and continuous learning: Generalization and Catastrophic Forgetting. We find that the overlap of task-specific neurons is strongly associated with generalization and specialization across tasks. Interestingly, at certain layers of LLMs, there is a high similarity in the parameters of different task-specific neurons, and such similarity is highly correlated with the generalization performance. Inspired by these findings, we propose a neuron-level continuous fine-tuning method that only fine-tunes the current task-specific neurons during continuous learning, and extensive experiments demonstrate the effectiveness of the proposed method. Our study provides insights into the interpretability of LLMs in multi-task learning.
♻ ☆ Balancing Class Accuracies for Language Models in Inference Time via Nonlinear Integer Programming
Large language models are good knowledge bases but struggle to perform equally well for all classes in simple text classification tasks. This paper investigates a fundamental inference-time problem in language models: imbalanced class accuracies. We find what's underneath the issue is a tendency to over-predict some classes while under-predicting some others, for both base LLMs and more sophisticatedly trained, larger chat LLMs. This class accuracy imbalance is difficult to solve from the root via ``better'' pre-training or fine-tuning strategies, but we show it can be effectively mitigated via inference-time optimization. To this end, we conceptualize and quantify the over- and under-prediction issue as the Contextual Oddity Bias (COBias), and propose the Debiasing as Nonlinear Integer Programming (DNIP) model to correct in-context learned class probabilities based on minimizing COBias and maximizing overall accuracy, without LLM parameter update. Considering that the DNIP model implicitly contains non-differentiable elements, we therefore use the simulated annealing algorithm to solve. We perform extensive evaluations on three LLMs across seven NLP classification tasks, in different prompting settings. Results show that DNIP simultaneously achieves significant COBias reduction (-27%) and accuracy improvement (+12%) over the conventional ICL approach, suggesting that inference-time mitigation of class accuracy imbalance is a promising direction to push forward LLM performances.
♻ ☆ Toward Adaptive Large Language Models Structured Pruning via Hybrid-grained Weight Importance Assessment AAAI 2025
Structured pruning for large language models (LLMs) has garnered significant academic interest due to its ability to efficiently compress and accelerate LLMs by eliminating redundant weight groups at a coarse-grained granularity. Current structured pruning methods for LLMs typically depend on a singular granularity for assessing weight importance, resulting in notable performance degradation in downstream tasks. Intriguingly, our empirical investigations reveal that utilizing unstructured pruning, which achieves better performance retention by pruning weights at a finer granularity, \emph{i.e.}, individual weights, yields significantly varied sparse LLM structures when juxtaposed to structured pruning. This suggests that evaluating both holistic and individual assessment for weight importance is essential for LLM pruning. Building on this insight, we introduce the Hybrid-grained Weight Importance Assessment (HyWIA), a novel method that merges fine-grained and coarse-grained evaluations of weight importance for the pruning of LLMs. Leveraging an attention mechanism, HyWIA adaptively determines the optimal blend of granularity in weight importance assessments in an end-to-end pruning manner. Extensive experiments on LLaMA-V1/V2, Vicuna, Baichuan, and Bloom across various benchmarks demonstrate the effectiveness of HyWIA in pruning LLMs. For example, HyWIA surpasses the cutting-edge LLM-Pruner by an average margin of 2.82% in accuracy across seven downstream tasks when pruning LLaMA-7B by 50%.
comment: AAAI 2025
♻ ☆ Knowledge in Superposition: Unveiling the Failures of Lifelong Knowledge Editing for Large Language Models AAAI 2025
Knowledge editing aims to update outdated or incorrect knowledge in large language models (LLMs). However, current knowledge editing methods have limited scalability for lifelong editing. This study explores the fundamental reason why knowledge editing fails in lifelong editing. We begin with the closed-form solution derived from linear associative memory, which underpins state-of-the-art knowledge editing methods. We extend the solution from single editing to lifelong editing, and through rigorous mathematical derivation, identify an interference term in the final solution, suggesting that editing knowledge may impact irrelevant knowledge. Further analysis of the interference term reveals a close relationship with superposition between knowledge representations. When knowledge superposition does not exist in language models, the interference term vanishes, allowing for lossless knowledge editing. Experiments across numerous language models reveal that knowledge superposition is universal, exhibiting high kurtosis, zero mean, and heavy-tailed distributions with clear scaling laws. Ultimately, by combining theory and experiments, we demonstrate that knowledge superposition is the fundamental reason for the failure of lifelong editing. Moreover, this is the first study to investigate knowledge editing from the perspective of superposition and provides a comprehensive observation of superposition across numerous real-world language models. Code available at https://github.com/ChenhuiHu/knowledge_in_superposition.
comment: To be published in AAAI 2025
♻ ☆ AI as Humanity's Salieri: Quantifying Linguistic Creativity of Language Models via Systematic Attribution of Machine Text against Web Text
Creativity has long been considered one of the most difficult aspect of human intelligence for AI to mimic. However, the rise of Large Language Models (LLMs), like ChatGPT, has raised questions about whether AI can match or even surpass human creativity. We present CREATIVITY INDEX as the first step to quantify the linguistic creativity of a text by reconstructing it from existing text snippets on the web. CREATIVITY INDEX is motivated by the hypothesis that the seemingly remarkable creativity of LLMs may be attributable in large part to the creativity of human-written texts on the web. To compute CREATIVITY INDEX efficiently, we introduce DJ SEARCH, a novel dynamic programming algorithm that can search verbatim and near-verbatim matches of text snippets from a given document against the web. Experiments reveal that the CREATIVITY INDEX of professional human authors is on average 66.2% higher than that of LLMs, and that alignment reduces the CREATIVITY INDEX of LLMs by an average of 30.1%. In addition, we find that distinguished authors like Hemingway exhibit measurably higher CREATIVITY INDEX compared to other human writers. Finally, we demonstrate that CREATIVITY INDEX can be used as a surprisingly effective criterion for zero-shot machine text detection, surpassing the strongest existing zero-shot system, DetectGPT, by a significant margin of 30.2%, and even outperforming the strongest supervised system, GhostBuster, in five out of six domains.
♻ ☆ On the Effectiveness of ASR Representations in Real-world Noisy Speech Emotion Recognition
This paper proposes an efficient attempt to noisy speech emotion recognition (NSER). Conventional NSER approaches have proven effective in mitigating the impact of artificial noise sources, such as white Gaussian noise, but are limited to non-stationary noises in real-world environments due to their complexity and uncertainty. To overcome this limitation, we introduce a new method for NSER by adopting the automatic speech recognition (ASR) model as a noise-robust feature extractor to eliminate non-vocal information in noisy speech. We first obtain intermediate layer information from the ASR model as a feature representation for emotional speech and then apply this representation for the downstream NSER task. Our experimental results show that 1) the proposed method achieves better NSER performance compared with the conventional noise reduction method, 2) outperforms self-supervised learning approaches, and 3) even outperforms text-based approaches using ASR transcription or the ground truth transcription of noisy speech.
comment: Submitted to IEEE/ACM Transactions on Audio, Speech, and Language Processing
♻ ☆ Towards Global AI Inclusivity: A Large-Scale Multilingual Terminology Dataset (GIST)
The field of machine translation has achieved significant advancements, yet domain-specific terminology translation, particularly in AI, remains challenging. We introduce GIST, a large-scale multilingual AI terminology dataset containing 5K terms extracted from top AI conference papers spanning 2000 to 2023. The terms are translated into Arabic, Chinese, French, Japanese, and Russian using a hybrid framework that combines LLMs for extraction with human expertise for translation. The dataset's quality is benchmarked against existing resources, demonstrating superior translation accuracy through crowdsourced evaluation. GIST is integrated into translation workflows using post-translation refinement methods that require no retraining, where LLM prompting consistently improves BLEU and COMET scores. A web demonstration on the ACL Anthology platform highlights its practical application, showcasing improved accessibility for non-English speakers. This work aims to address critical gaps in AI terminology resources and fosters global inclusivity and collaboration in AI research.
♻ ☆ LiveIdeaBench: Evaluating LLMs' Scientific Creativity and Idea Generation with Minimal Context
While Large Language Models (LLMs) have demonstrated remarkable capabilities in scientific tasks, existing evaluation frameworks primarily assess their performance using rich contextual inputs, overlooking their ability to generate novel ideas from minimal information. We introduce LiveIdeaBench, a comprehensive benchmark that evaluates LLMs' scientific creativity and divergent thinking capabilities using single-keyword prompts. Drawing from Guilford's creativity theory, our framework employs a dynamic panel of state-of-the-art LLMs to assess generated ideas across four key dimensions: originality, feasibility, fluency, and flexibility. Through extensive experimentation with 20 leading models across 1,180 keywords spanning 18 scientific domains, we reveal that scientific creative ability shows distinct patterns from general intelligence metrics. Notably, our results demonstrate that models like QwQ-32B-preview achieve comparable creative performance to top-tier models like o1-preview, despite significant gaps in their general intelligence scores. These findings highlight the importance of specialized evaluation frameworks for scientific creativity and suggest that the development of creative capabilities in LLMs may follow different trajectories than traditional problem-solving abilities.
comment: Updated author list, Corrected some issues and ref
Machine Learning 17
☆ Kolmogorov-Arnold Recurrent Network for Short Term Load Forecasting Across Diverse Consumers
Load forecasting plays a crucial role in energy management, directly impacting grid stability, operational efficiency, cost reduction, and environmental sustainability. Traditional Vanilla Recurrent Neural Networks (RNNs) face issues such as vanishing and exploding gradients, whereas sophisticated RNNs such as LSTMs have shown considerable success in this domain. However, these models often struggle to accurately capture complex and sudden variations in energy consumption, and their applicability is typically limited to specific consumer types, such as offices or schools. To address these challenges, this paper proposes the Kolmogorov-Arnold Recurrent Network (KARN), a novel load forecasting approach that combines the flexibility of Kolmogorov-Arnold Networks with RNN's temporal modeling capabilities. KARN utilizes learnable temporal spline functions and edge-based activations to better model non-linear relationships in load data, making it adaptable across a diverse range of consumer types. The proposed KARN model was rigorously evaluated on a variety of real-world datasets, including student residences, detached homes, a home with electric vehicle charging, a townhouse, and industrial buildings. Across all these consumer categories, KARN consistently outperformed traditional Vanilla RNNs, while it surpassed LSTM and Gated Recurrent Units (GRUs) in six buildings. The results demonstrate KARN's superior accuracy and applicability, making it a promising tool for enhancing load forecasting in diverse energy management scenarios.
☆ Compact Bayesian Neural Networks via pruned MCMC sampling
Bayesian Neural Networks (BNNs) offer robust uncertainty quantification in model predictions, but training them presents a significant computational challenge. This is mainly due to the problem of sampling multimodal posterior distributions using Markov Chain Monte Carlo (MCMC) sampling and variational inference algorithms. Moreover, the number of model parameters scales exponentially with additional hidden layers, neurons, and features in the dataset. Typically, a significant portion of these densely connected parameters are redundant and pruning a neural network not only improves portability but also has the potential for better generalisation capabilities. In this study, we address some of the challenges by leveraging MCMC sampling with network pruning to obtain compact probabilistic models having removed redundant parameters. We sample the posterior distribution of model parameters (weights and biases) and prune weights with low importance, resulting in a compact model. We ensure that the compact BNN retains its ability to estimate uncertainty via the posterior distribution while retaining the model training and generalisation performance accuracy by adapting post-pruning resampling. We evaluate the effectiveness of our MCMC pruning strategy on selected benchmark datasets for regression and classification problems through empirical result analysis. We also consider two coral reef drill-core lithology classification datasets to test the robustness of the pruning model in complex real-world datasets. We further investigate if refining compact BNN can retain any loss of performance. Our results demonstrate the feasibility of training and pruning BNNs using MCMC whilst retaining generalisation performance with over 75% reduction in network size. This paves the way for developing compact BNN models that provide uncertainty estimates for real-world applications.
comment: 22 pages, 11 figures
☆ Sanidha: A Studio Quality Multi-Modal Dataset for Carnatic Music
Music source separation demixes a piece of music into its individual sound sources (vocals, percussion, melodic instruments, etc.), a task with no simple mathematical solution. It requires deep learning methods involving training on large datasets of isolated music stems. The most commonly available datasets are made from commercial Western music, limiting the models' applications to non-Western genres like Carnatic music. Carnatic music is a live tradition, with the available multi-track recordings containing overlapping sounds and bleeds between the sources. This poses a challenge to commercially available source separation models like Spleeter and Hybrid Demucs. In this work, we introduce 'Sanidha', the first open-source novel dataset for Carnatic music, offering studio-quality, multi-track recordings with minimal to no overlap or bleed. Along with the audio files, we provide high-definition videos of the artists' performances. Additionally, we fine-tuned Spleeter, one of the most commonly used source separation models, on our dataset and observed improved SDR performance compared to fine-tuning on a pre-existing Carnatic multi-track dataset. The outputs of the fine-tuned model with 'Sanidha' are evaluated through a listening study.
comment: Accepted to the 25th International Society for Music Information Retrieval Conference (ISMIR 2024)
☆ A Hessian-informed hyperparameter optimization for differential learning rate
Differential learning rate (DLR), a technique that applies different learning rates to different model parameters, has been widely used in deep learning and achieved empirical success via its various forms. For example, parameter-efficient fine-tuning (PEFT) applies zero learning rates to most parameters so as to significantly save the computational cost. At the core, DLR leverages the observation that different parameters can have different loss curvature, which is hard to characterize in general. We propose the Hessian-informed differential learning rate (Hi-DLR), an efficient approach that solves the hyperparameter optimization (HPO) of learning rates and captures the loss curvature for any model and optimizer adaptively. Given a proper grouping of parameters, we empirically demonstrate that Hi-DLR can improve the convergence by dynamically determining the learning rates during the training. Furthermore, we can quantify the influence of different parameters and freeze the less-contributing parameters, which leads to a new PEFT that automatically adapts to various tasks and models. Additionally, Hi-DLR also exhibits comparable performance on various full model training tasks.
☆ Comparison of Autoencoders for tokenization of ASL datasets
Generative AI, powered by large language models (LLMs), has revolutionized applications across text, audio, images, and video. This study focuses on developing and evaluating encoder-decoder architectures for the American Sign Language (ASL) image dataset, consisting of 87,000 images across 29 hand sign classes. Three approaches were compared: Feedforward Autoencoders, Convolutional Autoencoders, and Diffusion Autoencoders. The Diffusion Autoencoder outperformed the others, achieving the lowest mean squared error (MSE) and highest Mean Opinion Score (MOS) due to its probabilistic noise modeling and iterative denoising capabilities. The Convolutional Autoencoder demonstrated effective spatial feature extraction but lacked the robustness of the diffusion process, while the Feedforward Autoencoder served as a baseline with limitations in handling complex image data. Objective and subjective evaluations confirmed the superiority of the Diffusion Autoencoder for high-fidelity image reconstruction, emphasizing its potential in multimodal AI applications such as sign language recognition and generation. This work provides critical insights into designing robust encoder-decoder systems to advance multimodal AI capabilities.
comment: 9 pages, 2 tables, 4 figures
☆ Super-Resolution of 3D Micro-CT Images Using Generative Adversarial Networks: Enhancing Resolution and Segmentation Accuracy
We develop a procedure for substantially improving the quality of segmented 3D micro-Computed Tomography (micro-CT) images of rocks with a Machine Learning (ML) Generative Model. The proposed model enhances the resolution eightfold (8x) and addresses segmentation inaccuracies due to the overlapping X-ray attenuation in micro-CT measurement for different rock minerals and phases. The proposed generative model is a 3D Deep Convolutional Wasserstein Generative Adversarial Network with Gradient Penalty (3D DC WGAN-GP). The algorithm is trained on segmented 3D low-resolution micro-CT images and segmented unpaired complementary 2D high-resolution Laser Scanning Microscope (LSM) images. The algorithm was demonstrated on multiple samples of Berea sandstones. We achieved high-quality super-resolved 3D images with a resolution of 0.4375 micro-m/voxel and accurate segmentation for constituting minerals and pore space. The described procedure can significantly expand the modern capabilities of digital rock physics.
comment: 24 pages, 9 figures
☆ A group-theoretic framework for machine learning in hyperbolic spaces
Embedding the data in hyperbolic spaces can preserve complex relationships in very few dimensions, thus enabling compact models and improving efficiency of machine learning (ML) algorithms. The underlying idea is that hyperbolic representations can prevent the loss of important structural information for certain ubiquitous types of data. However, further advances in hyperbolic ML require more principled mathematical approaches and adequate geometric methods. The present study aims at enhancing mathematical foundations of hyperbolic ML by combining group-theoretic and conformal-geometric arguments with optimization and statistical techniques. Precisely, we introduce the notion of the mean (barycenter) and the novel family of probability distributions on hyperbolic balls. We further propose efficient optimization algorithms for computation of the barycenter and for maximum likelihood estimation. One can build upon basic concepts presented here in order to design more demanding algorithms and implement hyperbolic deep learning pipelines.
comment: 22 pages, 4 figures
☆ Neural equilibria for long-term prediction of nonlinear conservation laws
We introduce Neural Discrete Equilibrium (NeurDE), a machine learning (ML) approach for long-term forecasting of flow phenomena that relies on a "lifting" of physical conservation laws into the framework of kinetic theory. The kinetic formulation provides an excellent structure for ML algorithms by separating nonlinear, non-local physics into a nonlinear but local relaxation to equilibrium and a linear non-local transport. This separation allows the ML to focus on the local nonlinear components while addressing the simpler linear transport with efficient classical numerical algorithms. To accomplish this, we design an operator network that maps macroscopic observables to equilibrium states in a manner that maximizes entropy, yielding expressive BGK-type collisions. By incorporating our surrogate equilibrium into the lattice Boltzmann (LB) algorithm, we achieve accurate flow forecasts for a wide range of challenging flows. We show that NeurDE enables accurate prediction of compressible flows, including supersonic flows, while tracking shocks over hundreds of time steps, using a small velocity lattice-a heretofore unattainable feat without expensive numerical root finding.
☆ Harnessing Large Language Models for Disaster Management: A Survey
Large language models (LLMs) have revolutionized scientific research with their exceptional capabilities and transformed various fields. Among their practical applications, LLMs have been playing a crucial role in mitigating threats to human life, infrastructure, and the environment. Despite growing research in disaster LLMs, there remains a lack of systematic review and in-depth analysis of LLMs for natural disaster management. To address the gap, this paper presents a comprehensive survey of existing LLMs in natural disaster management, along with a taxonomy that categorizes existing works based on disaster phases and application scenarios. By collecting public datasets and identifying key challenges and opportunities, this study aims to guide the professional community in developing advanced LLMs for disaster management to enhance the resilience against natural disasters.
☆ Automatic Double Reinforcement Learning in Semiparametric Markov Decision Processes with Applications to Long-Term Causal Inference
Double reinforcement learning (DRL) enables statistically efficient inference on the value of a policy in a nonparametric Markov Decision Process (MDP) given trajectories generated by another policy. However, this approach necessarily requires stringent overlap between the state distributions, which is often violated in practice. To relax this requirement and extend DRL, we study efficient inference on linear functionals of the $Q$-function (of which policy value is a special case) in infinite-horizon, time-invariant MDPs under semiparametric restrictions on the $Q$-function. These restrictions can reduce the overlap requirement and lower the efficiency bound, yielding more precise estimates. As an important example, we study the evaluation of long-term value under domain adaptation, given a few short trajectories from the new domain and restrictions on the difference between the domains. This can be used for long-term causal inference. Our method combines flexible estimates of the $Q$-function and the Riesz representer of the functional of interest (e.g., the stationary state density ratio for policy value) and is automatic in that we do not need to know the form of the latter - only the functional we care about. To address potential model misspecification bias, we extend the adaptive debiased machine learning (ADML) framework of \citet{van2023adaptive} to construct nonparametrically valid and superefficient estimators that adapt to the functional form of the $Q$-function. As a special case, we propose a novel adaptive debiased plug-in estimator that uses isotonic-calibrated fitted $Q$-iteration - a new calibration algorithm for MDPs - to circumvent the computational challenges of estimating debiasing nuisances from min-max objectives.
☆ A Hybrid Virtual Element Method and Deep Learning Approach for Solving One-Dimensional Euler-Bernoulli Beams
A hybrid framework integrating the Virtual Element Method (VEM) with deep learning is presented as an initial step toward developing efficient and flexible numerical models for one-dimensional Euler-Bernoulli beams. The primary aim is to explore a data-driven surrogate model capable of predicting displacement fields across varying material and geometric parameters while maintaining computational efficiency. Building upon VEM's ability to handle higher-order polynomials and non-conforming discretizations, the method offers a robust numerical foundation for structural mechanics. A neural network architecture is introduced to separately process nodal and material-specific data, effectively capturing complex interactions with minimal reliance on large datasets. To address challenges in training, the model incorporates Sobolev training and GradNorm techniques, ensuring balanced loss contributions and enhanced generalization. While this framework is in its early stages, it demonstrates the potential for further refinement and development into a scalable alternative to traditional methods. The proposed approach lays the groundwork for advancing numerical and data-driven techniques in beam modeling, offering a foundation for future research in structural mechanics.
☆ Optimal Online Bookmaking for Binary Games
In online betting, the bookmaker can update the payoffs it offers on a particular event many times before the event takes place, and the updated payoffs may depend on the bets accumulated thus far. We study the problem of bookmaking with the goal of maximizing the return in the worst-case, with respect to the gamblers' behavior and the event's outcome. We formalize this problem as the \emph{Optimal Online Bookmaking game}, and provide the exact solution for the binary case. To this end, we develop the optimal bookmaking strategy, which relies on a new technique called bi-balancing trees, that assures that the house loss is the same for all \emph{decisive} betting sequences, where the gambler bets all its money on a single outcome in each round.
♻ ☆ NeuJeans: Private Neural Network Inference with Joint Optimization of Convolution and FHE Bootstrapping
Fully homomorphic encryption (FHE) is a promising cryptographic primitive for realizing private neural network inference (PI) services by allowing a client to fully offload the inference task to a cloud server while keeping the client data oblivious to the server. This work proposes NeuJeans, an FHE-based solution for the PI of deep convolutional neural networks (CNNs). NeuJeans tackles the critical problem of the enormous computational cost for the FHE evaluation of CNNs. We introduce a novel encoding method called Coefficients-in-Slot (CinS) encoding, which enables multiple convolutions in one HE multiplication without costly slot permutations. We further observe that CinS encoding is obtained by conducting the first several steps of the Discrete Fourier Transform (DFT) on a ciphertext in conventional Slot encoding. This property enables us to save the conversion between CinS and Slot encodings as bootstrapping a ciphertext starts with DFT. Exploiting this, we devise optimized execution flows for various two-dimensional convolution (conv2d) operations and apply them to end-to-end CNN implementations. NeuJeans accelerates the performance of conv2d-activation sequences by up to 5.68 times compared to state-of-the-art FHE-based PI work and performs the PI of a CNN at the scale of ImageNet within a mere few seconds.
comment: 15 pages, 6 figures, published at ACM 2024
♻ ☆ Parallel Momentum Methods Under Biased Gradient Estimations
Parallel stochastic gradient methods are gaining prominence in solving large-scale machine learning problems that involve data distributed across multiple nodes. However, obtaining unbiased stochastic gradients, which have been the focus of most theoretical research, is challenging in many distributed machine learning applications. The gradient estimations easily become biased, for example, when gradients are compressed or clipped, when data is shuffled, and in meta-learning and reinforcement learning. In this work, we establish worst-case bounds on parallel momentum methods under biased gradient estimation on both general non-convex and $\mu$-PL problems. Our analysis covers general distributed optimization problems, and we work out the implications for special cases where gradient estimates are biased, i.e. in meta-learning and when the gradients are compressed or clipped. Our numerical experiments verify our theoretical findings and show faster convergence performance of momentum methods than traditional biased gradient descent.
comment: 12 pages
♻ ☆ Multi-channel learning for integrating structural hierarchies into context-dependent molecular representation
Reliable molecular property prediction is essential for various scientific endeavors and industrial applications, such as drug discovery. However, the data scarcity, combined with the highly non-linear causal relationships between physicochemical and biological properties and conventional molecular featurization schemes, complicates the development of robust molecular machine learning models. Self-supervised learning (SSL) has emerged as a popular solution, utilizing large-scale, unannotated molecular data to learn a foundational representation of chemical space that might be advantageous for downstream tasks. Yet, existing molecular SSL methods largely overlook chemical knowledge, including molecular structure similarity, scaffold composition, and the context-dependent aspects of molecular properties when operating over the chemical space. They also struggle to learn the subtle variations in structure-activity relationship. This paper introduces a novel pre-training framework that learns robust and generalizable chemical knowledge. It leverages the structural hierarchy within the molecule, embeds them through distinct pre-training tasks across channels, and aggregates channel information in a task-specific manner during fine-tuning. Our approach demonstrates competitive performance across various molecular property benchmarks and offers strong advantages in particularly challenging yet ubiquitous scenarios like activity cliffs.
♻ ☆ ChatGPT Needs SPADE (Sustainability, PrivAcy, Digital divide, and Ethics) Evaluation: A Review
ChatGPT is another large language model (LLM) vastly available for the consumers on their devices but due to its performance and ability to converse effectively, it has gained a huge popularity amongst research as well as industrial community. Recently, many studies have been published to show the effectiveness, efficiency, integration, and sentiments of chatGPT and other LLMs. In contrast, this study focuses on the important aspects that are mostly overlooked, i.e. sustainability, privacy, digital divide, and ethics and suggests that not only chatGPT but every subsequent entry in the category of conversational bots should undergo Sustainability, PrivAcy, Digital divide, and Ethics (SPADE) evaluation. This paper discusses in detail the issues and concerns raised over chatGPT in line with aforementioned characteristics. We also discuss the recent EU AI Act briefly in accordance with the SPADE evaluation. We support our hypothesis by some preliminary data collection and visualizations along with hypothesized facts. We also suggest mitigations and recommendations for each of the concerns. Furthermore, we also suggest some policies and recommendations for EU AI policy act concerning ethics, digital divide, and sustainability
comment: 29 pages, 8 figures, 4 tables
♻ ☆ A Survey on Reinforcement Learning Applications in SLAM
The emergence of mobile robotics, particularly in the automotive industry, introduces a promising era of enriched user experiences and adept handling of complex navigation challenges. The realization of these advancements necessitates a focused technological effort and the successful execution of numerous intricate tasks, particularly in the critical domain of Simultaneous Localization and Mapping (SLAM). Various artificial intelligence (AI) methodologies, such as deep learning and reinforcement learning, present viable solutions to address the challenges in SLAM. This study specifically explores the application of reinforcement learning in the context of SLAM. By enabling the agent (the robot) to iteratively interact with and receive feedback from its environment, reinforcement learning facilitates the acquisition of navigation and mapping skills, thereby enhancing the robot's decision-making capabilities. This approach offers several advantages, including improved navigation proficiency, increased resilience, reduced dependence on sensor precision, and refinement of the decision-making process. The findings of this study, which provide an overview of reinforcement learning's utilization in SLAM, reveal significant advancements in the field. The investigation also highlights the evolution and innovative integration of these techniques.
Artificial Intelligence 25
☆ Kolmogorov-Arnold Recurrent Network for Short Term Load Forecasting Across Diverse Consumers
Load forecasting plays a crucial role in energy management, directly impacting grid stability, operational efficiency, cost reduction, and environmental sustainability. Traditional Vanilla Recurrent Neural Networks (RNNs) face issues such as vanishing and exploding gradients, whereas sophisticated RNNs such as LSTMs have shown considerable success in this domain. However, these models often struggle to accurately capture complex and sudden variations in energy consumption, and their applicability is typically limited to specific consumer types, such as offices or schools. To address these challenges, this paper proposes the Kolmogorov-Arnold Recurrent Network (KARN), a novel load forecasting approach that combines the flexibility of Kolmogorov-Arnold Networks with RNN's temporal modeling capabilities. KARN utilizes learnable temporal spline functions and edge-based activations to better model non-linear relationships in load data, making it adaptable across a diverse range of consumer types. The proposed KARN model was rigorously evaluated on a variety of real-world datasets, including student residences, detached homes, a home with electric vehicle charging, a townhouse, and industrial buildings. Across all these consumer categories, KARN consistently outperformed traditional Vanilla RNNs, while it surpassed LSTM and Gated Recurrent Units (GRUs) in six buildings. The results demonstrate KARN's superior accuracy and applicability, making it a promising tool for enhancing load forecasting in diverse energy management scenarios.
☆ Enhancing Patient-Centric Communication: Leveraging LLMs to Simulate Patient Perspectives
Large Language Models (LLMs) have demonstrated impressive capabilities in role-playing scenarios, particularly in simulating domain-specific experts using tailored prompts. This ability enables LLMs to adopt the persona of individuals with specific backgrounds, offering a cost-effective and efficient alternative to traditional, resource-intensive user studies. By mimicking human behavior, LLMs can anticipate responses based on concrete demographic or professional profiles. In this paper, we evaluate the effectiveness of LLMs in simulating individuals with diverse backgrounds and analyze the consistency of these simulated behaviors compared to real-world outcomes. In particular, we explore the potential of LLMs to interpret and respond to discharge summaries provided to patients leaving the Intensive Care Unit (ICU). We evaluate and compare with human responses the comprehensibility of discharge summaries among individuals with varying educational backgrounds, using this analysis to assess the strengths and limitations of LLM-driven simulations. Notably, when LLMs are primed with educational background information, they deliver accurate and actionable medical guidance 88% of the time. However, when other information is provided, performance significantly drops, falling below random chance levels. This preliminary study shows the potential benefits and pitfalls of automatically generating patient-specific health information from diverse populations. While LLMs show promise in simulating health personas, our results highlight critical gaps that must be addressed before they can be reliably used in clinical settings. Our findings suggest that a straightforward query-response model could outperform a more tailored approach in delivering health information. This is a crucial first step in understanding how LLMs can be optimized for personalized health communication while maintaining accuracy.
☆ Generative Artificial Intelligence-Supported Pentesting: A Comparison between Claude Opus, GPT-4, and Copilot
The advent of Generative Artificial Intelligence (GenAI) has brought a significant change to our society. GenAI can be applied across numerous fields, with particular relevance in cybersecurity. Among the various areas of application, its use in penetration testing (pentesting) or ethical hacking processes is of special interest. In this paper, we have analyzed the potential of leading generic-purpose GenAI tools-Claude Opus, GPT-4 from ChatGPT, and Copilot-in augmenting the penetration testing process as defined by the Penetration Testing Execution Standard (PTES). Our analysis involved evaluating each tool across all PTES phases within a controlled virtualized environment. The findings reveal that, while these tools cannot fully automate the pentesting process, they provide substantial support by enhancing efficiency and effectiveness in specific tasks. Notably, all tools demonstrated utility; however, Claude Opus consistently outperformed the others in our experimental scenarios.
☆ Compact Bayesian Neural Networks via pruned MCMC sampling
Bayesian Neural Networks (BNNs) offer robust uncertainty quantification in model predictions, but training them presents a significant computational challenge. This is mainly due to the problem of sampling multimodal posterior distributions using Markov Chain Monte Carlo (MCMC) sampling and variational inference algorithms. Moreover, the number of model parameters scales exponentially with additional hidden layers, neurons, and features in the dataset. Typically, a significant portion of these densely connected parameters are redundant and pruning a neural network not only improves portability but also has the potential for better generalisation capabilities. In this study, we address some of the challenges by leveraging MCMC sampling with network pruning to obtain compact probabilistic models having removed redundant parameters. We sample the posterior distribution of model parameters (weights and biases) and prune weights with low importance, resulting in a compact model. We ensure that the compact BNN retains its ability to estimate uncertainty via the posterior distribution while retaining the model training and generalisation performance accuracy by adapting post-pruning resampling. We evaluate the effectiveness of our MCMC pruning strategy on selected benchmark datasets for regression and classification problems through empirical result analysis. We also consider two coral reef drill-core lithology classification datasets to test the robustness of the pruning model in complex real-world datasets. We further investigate if refining compact BNN can retain any loss of performance. Our results demonstrate the feasibility of training and pruning BNNs using MCMC whilst retaining generalisation performance with over 75% reduction in network size. This paves the way for developing compact BNN models that provide uncertainty estimates for real-world applications.
comment: 22 pages, 11 figures
☆ Patent Novelty Assessment Accelerating Innovation and Patent Prosecution
In the rapidly evolving landscape of technological innovation, safeguarding intellectual property rights through patents is crucial for fostering progress and stimulating research and development investments. This report introduces a ground-breaking Patent Novelty Assessment and Claim Generation System, meticulously crafted to dissect the inventive aspects of intellectual property and simplify access to extensive patent claim data. Addressing a crucial gap in academic institutions, our system provides college students and researchers with an intuitive platform to navigate and grasp the intricacies of patent claims, particularly tailored for the nuances of Chinese patents. Unlike conventional analysis systems, our initiative harnesses a proprietary Chinese API to ensure unparalleled precision and relevance. The primary challenge lies in the complexity of accessing and comprehending diverse patent claims, inhibiting effective innovation upon existing ideas. Our solution aims to overcome these barriers by offering a bespoke approach that seamlessly retrieves comprehensive claim information, finely tuned to the specifics of the Chinese patent landscape. By equipping users with efficient access to comprehensive patent claim information, our transformative platform seeks to ignite informed exploration and innovation in the ever-evolving domain of intellectual property. Its envisioned impact transcends individual colleges, nurturing an environment conducive to research and development while deepening the understanding of patented concepts within the academic community.
☆ The Einstein Test: Towards a Practical Test of a Machine's Ability to Exhibit Superintelligence
Creative and disruptive insights (CDIs), such as the development of the theory of relativity, have punctuated human history, marking pivotal shifts in our intellectual trajectory. Recent advancements in artificial intelligence (AI) have sparked debates over whether state of the art models possess the capacity to generate CDIs. We argue that the ability to create CDIs should be regarded as a significant feature of machine superintelligence (SI).To this end, we propose a practical test to evaluate whether an approach to AI targeting SI can yield novel insights of this kind. We propose the Einstein test: given the data available prior to the emergence of a known CDI, can an AI independently reproduce that insight (or one that is formally equivalent)? By achieving such a milestone, a machine can be considered to at least match humanity's past top intellectual achievements, and therefore to have the potential to surpass them.
☆ An Empirical Study of Deep Reinforcement Learning in Continuing Tasks
In reinforcement learning (RL), continuing tasks refer to tasks where the agent-environment interaction is ongoing and can not be broken down into episodes. These tasks are suitable when environment resets are unavailable, agent-controlled, or predefined but where all rewards-including those beyond resets-are critical. These scenarios frequently occur in real-world applications and can not be modeled by episodic tasks. While modern deep RL algorithms have been extensively studied and well understood in episodic tasks, their behavior in continuing tasks remains underexplored. To address this gap, we provide an empirical study of several well-known deep RL algorithms using a suite of continuing task testbeds based on Mujoco and Atari environments, highlighting several key insights concerning continuing tasks. Using these testbeds, we also investigate the effectiveness of a method for improving temporal-difference-based RL algorithms in continuing tasks by centering rewards, as introduced by Naik et al. (2024). While their work primarily focused on this method in conjunction with Q-learning, our results extend their findings by demonstrating that this method is effective across a broader range of algorithms, scales to larger tasks, and outperforms two other reward-centering approaches.
☆ Why are we living the age of AI applications right now? The long innovation path from AI's birth to a child's bedtime magic
Today a four-year-old child who does not know how to read or write can now create bedtime stories with graphical illustrations and narrated audio, using AI tools that seamlessly transform speech into text, generate visuals, and convert text back into speech in a natural and engaging manner. This remarkable example demonstrates why we are living in the age of AI applications. This paper examines contemporary leading AI applications and traces their historical development, highlighting the major advancements that have enabled their realization. Five key factors are identified: 1) The evolution of computational hardware (CPUs and GPUs), enabling the training of complex AI models 2) The vast digital archives provided by the World Wide Web, which serve as a foundational data resource for AI systems 3) The ubiquity of mobile computing, with smartphones acting as powerful, accessible small computers in the hands of billions 4) The rise of industrial-scale cloud infrastructures, offering elastic computational power for AI training and deployment 5) Breakthroughs in AI research, including neural networks, backpropagation, and the "Attention is All You Need" framework, which underpin modern AI capabilities. These innovations have elevated AI from solving narrow tasks to enabling applications like ChatGPT that are adaptable for numerous use cases, redefining human-computer interaction. By situating these developments within a historical context, the paper highlights the critical milestones that have made AI's current capabilities both possible and widely accessible, offering profound implications for society.
comment: 14 pages, 8 figures
☆ Risk-Averse Finetuning of Large Language Models
We consider the challenge of mitigating the generation of negative or toxic content by the Large Language Models (LLMs) in response to certain prompts. We propose integrating risk-averse principles into LLM fine-tuning to minimize the occurrence of harmful outputs, particularly rare but significant events. By optimizing the risk measure of Conditional Value at Risk (CVaR), our methodology trains LLMs to exhibit superior performance in avoiding toxic outputs while maintaining effectiveness in generative tasks. Empirical evaluations on sentiment modification and toxicity mitigation tasks demonstrate the efficacy of risk-averse reinforcement learning with human feedback (RLHF) in promoting a safer and more constructive online discourse environment.
comment: Neurips 2024
☆ MedGrad E-CLIP: Enhancing Trust and Transparency in AI-Driven Skin Lesion Diagnosis WACV
As deep learning models gain attraction in medical data, ensuring transparent and trustworthy decision-making is essential. In skin cancer diagnosis, while advancements in lesion detection and classification have improved accuracy, the black-box nature of these methods poses challenges in understanding their decision processes, leading to trust issues among physicians. This study leverages the CLIP (Contrastive Language-Image Pretraining) model, trained on different skin lesion datasets, to capture meaningful relationships between visual features and diagnostic criteria terms. To further enhance transparency, we propose a method called MedGrad E-CLIP, which builds on gradient-based E-CLIP by incorporating a weighted entropy mechanism designed for complex medical imaging like skin lesions. This approach highlights critical image regions linked to specific diagnostic descriptions. The developed integrated pipeline not only classifies skin lesions by matching corresponding descriptions but also adds an essential layer of explainability developed especially for medical data. By visually explaining how different features in an image relates to diagnostic criteria, this approach demonstrates the potential of advanced vision-language models in medical image analysis, ultimately improving transparency, robustness, and trust in AI-driven diagnostic systems.
comment: Accepted to 2025 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW)
☆ Defect Detection Network In PCB Circuit Devices Based on GAN Enhanced YOLOv11
This study proposes an advanced method for surface defect detection in printed circuit boards (PCBs) using an improved YOLOv11 model enhanced with a generative adversarial network (GAN). The approach focuses on identifying six common defect types: missing hole, rat bite, open circuit, short circuit, burr, and virtual welding. By employing GAN to generate synthetic defect images, the dataset is augmented with diverse and realistic patterns, improving the model's ability to generalize, particularly for complex and infrequent defects like burrs. The enhanced YOLOv11 model is evaluated on a PCB defect dataset, demonstrating significant improvements in accuracy, recall, and robustness, especially when dealing with defects in complex environments or small targets. This research contributes to the broader field of electronic design automation (EDA), where efficient defect detection is a crucial step in ensuring high-quality PCB manufacturing. By integrating advanced deep learning techniques, this approach enhances the automation and precision of defect detection, reducing reliance on manual inspection and accelerating design-to-production workflows. The findings underscore the importance of incorporating GAN-based data augmentation and optimized detection architectures in EDA processes, providing valuable insights for improving reliability and efficiency in PCB defect detection within industrial applications.
☆ A Foundational Generative Model for Breast Ultrasound Image Analysis
Foundational models have emerged as powerful tools for addressing various tasks in clinical settings. However, their potential development to breast ultrasound analysis remains untapped. In this paper, we present BUSGen, the first foundational generative model specifically designed for breast ultrasound image analysis. Pretrained on over 3.5 million breast ultrasound images, BUSGen has acquired extensive knowledge of breast structures, pathological features, and clinical variations. With few-shot adaptation, BUSGen can generate repositories of realistic and informative task-specific data, facilitating the development of models for a wide range of downstream tasks. Extensive experiments highlight BUSGen's exceptional adaptability, significantly exceeding real-data-trained foundational models in breast cancer screening, diagnosis, and prognosis. In breast cancer early diagnosis, our approach outperformed all board-certified radiologists (n=9), achieving an average sensitivity improvement of 16.5% (P-value<0.0001). Additionally, we characterized the scaling effect of using generated data which was as effective as the collected real-world data for training diagnostic models. Moreover, extensive experiments demonstrated that our approach improved the generalization ability of downstream models. Importantly, BUSGen protected patient privacy by enabling fully de-identified data sharing, making progress forward in secure medical data utilization. An online demo of BUSGen is available at https://aibus.bio.
comment: Peking University; Stanford University; Peking University Cancer Hospital & Institute; Peking Union Medical College Hospital; Cancer Hospital, Chinese Academy of Medical Sciences
☆ Transfer Learning of Tabular Data by Finetuning Large Language Models
Despite the artificial intelligence (AI) revolution, deep learning has yet to achieve much success with tabular data due to heterogeneous feature space and limited sample sizes without viable transfer learning. The new era of generative AI, powered by large language models (LLM), brings unprecedented learning opportunities to diverse data and domains. This paper investigates the effectiveness of an LLM application programming interface (API) and transfer learning of LLM in tabular data classification. LLM APIs respond to input text prompts with tokenized data and instructions, whereas transfer learning finetunes an LLM for a target classification task. This paper proposes an end-to-end finetuning of LLM to demonstrate cross-data transfer learning on ten benchmark data sets when large pre-trained tabular data models do not exist to facilitate transfer learning. The proposed LLM finetuning method outperforms state-of-the-art machine and deep learning methods on tabular data with less than ten features - a standard feature size for tabular data sets. The transfer learning approach uses a fraction of the computational cost of other deep learning or API-based solutions while ensuring competitive or superior classification performance.
☆ LarvSeg: Exploring Image Classification Data For Large Vocabulary Semantic Segmentation via Category-wise Attentive Classifier
Scaling up the vocabulary of semantic segmentation models is extremely challenging because annotating large-scale mask labels is labour-intensive and time-consuming. Recently, language-guided segmentation models have been proposed to address this challenge. However, their performance drops significantly when applied to out-of-distribution categories. In this paper, we propose a new large vocabulary semantic segmentation framework, called LarvSeg. Different from previous works, LarvSeg leverages image classification data to scale the vocabulary of semantic segmentation models as large-vocabulary classification datasets usually contain balanced categories and are much easier to obtain. However, for classification tasks, the category is image-level, while for segmentation we need to predict the label at pixel level. To address this issue, we first propose a general baseline framework to incorporate image-level supervision into the training process of a pixel-level segmentation model, making the trained network perform semantic segmentation on newly introduced categories in the classification data. We then observe that a model trained on segmentation data can group pixel features of categories beyond the training vocabulary. Inspired by this finding, we design a category-wise attentive classifier to apply supervision to the precise regions of corresponding categories to improve the model performance. Extensive experiments demonstrate that LarvSeg significantly improves the large vocabulary semantic segmentation performance, especially in the categories without mask labels. For the first time, we provide a 21K-category semantic segmentation model with the help of ImageNet21K. The code is available at https://github.com/HaojunYu1998/large_voc_seg.
comment: PRCV 2024
☆ A Comprehensive Evaluation of Large Language Models on Mental Illnesses in Arabic Context
Mental health disorders pose a growing public health concern in the Arab world, emphasizing the need for accessible diagnostic and intervention tools. Large language models (LLMs) offer a promising approach, but their application in Arabic contexts faces challenges including limited labeled datasets, linguistic complexity, and translation biases. This study comprehensively evaluates 8 LLMs, including general multi-lingual models, as well as bi-lingual ones, on diverse mental health datasets (such as AraDepSu, Dreaddit, MedMCQA), investigating the impact of prompt design, language configuration (native Arabic vs. translated English, and vice versa), and few-shot prompting on diagnostic performance. We find that prompt engineering significantly influences LLM scores mainly due to reduced instruction following, with our structured prompt outperforming a less structured variant on multi-class datasets, with an average difference of 14.5\%. While language influence on performance was modest, model selection proved crucial: Phi-3.5 MoE excelled in balanced accuracy, particularly for binary classification, while Mistral NeMo showed superior performance in mean absolute error for severity prediction tasks. Few-shot prompting consistently improved performance, with particularly substantial gains observed for GPT-4o Mini on multi-class classification, boosting accuracy by an average factor of 1.58. These findings underscore the importance of prompt optimization, multilingual analysis, and few-shot learning for developing culturally sensitive and effective LLM-based mental health tools for Arabic-speaking populations.
☆ What Is a Counterfactual Cause in Action Theories? AAMAS 2025
Since the proposal by Halpern and Pearl, reasoning about actual causality has gained increasing attention in artificial intelligence, ranging from domains such as model-checking and verification to reasoning about actions and knowledge. More recently, Batusov and Soutchanski proposed a notion of actual achievement cause in the situation calculus, amongst others, they can determine the cause of quantified effects in a given action history. While intuitively appealing, this notion of cause is not defined in a counterfactual perspective. In this paper, we propose a notion of cause based on counterfactual analysis. In the context of action history, we show that our notion of cause generalizes naturally to a notion of achievement cause. We analyze the relationship between our notion of the achievement cause and the achievement cause by Batusov and Soutchanski. Finally, we relate our account of cause to Halpern and Pearl's account of actual causality. Particularly, we note some nuances in applying a counterfactual viewpoint to disjunctive goals, a common thorn to definitions of actual causes.
comment: This is an extended report of our short paper accepted at AAMAS 2025
☆ SPAM: Spike-Aware Adam with Momentum Reset for Stable LLM Training
Large Language Models (LLMs) have demonstrated exceptional performance across diverse tasks, yet their training remains highly resource-intensive and susceptible to critical challenges such as training instability. A predominant source of this instability stems from gradient and loss spikes, which disrupt the learning process, often leading to costly interventions like checkpoint recovery and experiment restarts, further amplifying inefficiencies. This paper presents a comprehensive investigation into gradient spikes observed during LLM training, revealing their prevalence across multiple architectures and datasets. Our analysis shows that these spikes can be up to $1000\times$ larger than typical gradients, substantially deteriorating model performance. To address this issue, we propose Spike-Aware Adam with Momentum Reset SPAM, a novel optimizer designed to counteract gradient spikes through momentum reset and spike-aware gradient clipping. Extensive experiments, including both pre-training and fine-tuning, demonstrate that SPAM consistently surpasses Adam and its variants across various tasks, including (1) LLM pre-training from 60M to 1B, (2) 4-bit LLM pre-training,(3) reinforcement learning, and (4) Time Series Forecasting. Additionally, SPAM facilitates memory-efficient training by enabling sparse momentum, where only a subset of momentum terms are maintained and updated. When operating under memory constraints, SPAM outperforms state-of-the-art memory-efficient optimizers such as GaLore and Adam-Mini. Our work underscores the importance of mitigating gradient spikes in LLM training and introduces an effective optimization strategy that enhances both training stability and resource efficiency at scale. Code is available at https://github.com/TianjinYellow/SPAM-Optimizer.git
☆ An efficient approach to represent enterprise web application structure using Large Language Model in the service of Intelligent Quality Engineering
This paper presents a novel approach to represent enterprise web application structures using Large Language Models (LLMs) to enable intelligent quality engineering at scale. We introduce a hierarchical representation methodology that optimizes the few-shot learning capabilities of LLMs while preserving the complex relationships and interactions within web applications. The approach encompasses five key phases: comprehensive DOM analysis, multi-page synthesis, test suite generation, execution, and result analysis. Our methodology addresses existing challenges around usage of Generative AI techniques in automated software testing by developing a structured format that enables LLMs to understand web application architecture through in-context learning. We evaluated our approach using two distinct web applications: an e-commerce platform (Swag Labs) and a healthcare application (MediBox) which is deployed within Atalgo engineering environment. The results demonstrate success rates of 90\% and 70\%, respectively, in achieving automated testing, with high relevance scores for test cases across multiple evaluation criteria. The findings suggest that our representation approach significantly enhances LLMs' ability to generate contextually relevant test cases and provide better quality assurance overall, while reducing the time and effort required for testing.
comment: 16 pages, 1 figure and 4 tables, relevant for Gen AI and enterprise AI use cases
♻ ☆ Artificial Intelligence for Cochlear Implants: Review of Strategies, Challenges, and Perspectives
Automatic speech recognition (ASR) plays a pivotal role in our daily lives, offering utility not only for interacting with machines but also for facilitating communication for individuals with partial or profound hearing impairments. The process involves receiving the speech signal in analog form, followed by various signal processing algorithms to make it compatible with devices of limited capacities, such as cochlear implants (CIs). Unfortunately, these implants, equipped with a finite number of electrodes, often result in speech distortion during synthesis. Despite efforts by researchers to enhance received speech quality using various state-of-the-art (SOTA) signal processing techniques, challenges persist, especially in scenarios involving multiple sources of speech, environmental noise, and other adverse conditions. The advent of new artificial intelligence (AI) methods has ushered in cutting-edge strategies to address the limitations and difficulties associated with traditional signal processing techniques dedicated to CIs. This review aims to comprehensively cover advancements in CI-based ASR and speech enhancement, among other related aspects. The primary objective is to provide a thorough overview of metrics and datasets, exploring the capabilities of AI algorithms in this biomedical field, and summarizing and commenting on the best results obtained. Additionally, the review will delve into potential applications and suggest future directions to bridge existing research gaps in this domain.
♻ ☆ ChatGPT Needs SPADE (Sustainability, PrivAcy, Digital divide, and Ethics) Evaluation: A Review
ChatGPT is another large language model (LLM) vastly available for the consumers on their devices but due to its performance and ability to converse effectively, it has gained a huge popularity amongst research as well as industrial community. Recently, many studies have been published to show the effectiveness, efficiency, integration, and sentiments of chatGPT and other LLMs. In contrast, this study focuses on the important aspects that are mostly overlooked, i.e. sustainability, privacy, digital divide, and ethics and suggests that not only chatGPT but every subsequent entry in the category of conversational bots should undergo Sustainability, PrivAcy, Digital divide, and Ethics (SPADE) evaluation. This paper discusses in detail the issues and concerns raised over chatGPT in line with aforementioned characteristics. We also discuss the recent EU AI Act briefly in accordance with the SPADE evaluation. We support our hypothesis by some preliminary data collection and visualizations along with hypothesized facts. We also suggest mitigations and recommendations for each of the concerns. Furthermore, we also suggest some policies and recommendations for EU AI policy act concerning ethics, digital divide, and sustainability
comment: 29 pages, 8 figures, 4 tables
♻ ☆ Beyond Sight: Finetuning Generalist Robot Policies with Heterogeneous Sensors via Language Grounding
Interacting with the world is a multi-sensory experience: achieving effective general-purpose interaction requires making use of all available modalities -- including vision, touch, and audio -- to fill in gaps from partial observation. For example, when vision is occluded reaching into a bag, a robot should rely on its senses of touch and sound. However, state-of-the-art generalist robot policies are typically trained on large datasets to predict robot actions solely from visual and proprioceptive observations. In this work, we propose FuSe, a novel approach that enables finetuning visuomotor generalist policies on heterogeneous sensor modalities for which large datasets are not readily available by leveraging natural language as a common cross-modal grounding. We combine a multimodal contrastive loss with a sensory-grounded language generation loss to encode high-level semantics. In the context of robot manipulation, we show that FuSe enables performing challenging tasks that require reasoning jointly over modalities such as vision, touch, and sound in a zero-shot setting, such as multimodal prompting, compositional cross-modal prompting, and descriptions of objects it interacts with. We show that the same recipe is applicable to widely different generalist policies, including both diffusion-based generalist policies and large vision-language-action (VLA) models. Extensive experiments in the real world show that FuSeis able to increase success rates by over 20% compared to all considered baselines.
♻ ☆ Explaining Datasets in Words: Statistical Models with Natural Language Parameters
To make sense of massive data, we often fit simplified models and then interpret the parameters; for example, we cluster the text embeddings and then interpret the mean parameters of each cluster. However, these parameters are often high-dimensional and hard to interpret. To make model parameters directly interpretable, we introduce a family of statistical models -- including clustering, time series, and classification models -- parameterized by natural language predicates. For example, a cluster of text about COVID could be parameterized by the predicate "discusses COVID". To learn these statistical models effectively, we develop a model-agnostic algorithm that optimizes continuous relaxations of predicate parameters with gradient descent and discretizes them by prompting language models (LMs). Finally, we apply our framework to a wide range of problems: taxonomizing user chat dialogues, characterizing how they evolve across time, finding categories where one language model is better than the other, clustering math problems based on subareas, and explaining visual features in memorable images. Our framework is highly versatile, applicable to both textual and visual domains, can be easily steered to focus on specific properties (e.g. subareas), and explains sophisticated concepts that classical methods (e.g. n-gram analysis) struggle to produce.
♻ ☆ Unlocking the Power of LLM Uncertainty for Active In-Context Example Selection
Large Language Models (LLMs) have shown remarkable performance across a wide range of downstream tasks. However, it is challenging for users to discern whether the responses of LLM are generated with certainty or are fabricated to meet user expectations. In this paper, we introduce Uncertainty Tripartite Testing Paradigm (Unc-TTP), a novel method for classifying LLM uncertainty by leveraging output inconsistency. Specifically, Unc-TTP performs three rounds of sampling under varying label injection interference, enumerating all possible outcomes, and uses the degree of output inconsistency as the indicator of the LLM's intrinsic uncertainty. To validate the effectiveness of this inconsistency-defined uncertainty, we draw inspiration from Active Learning, comparing the informativeness of actively selected in-context examples. Our experiments show that uncertainty examples selected via Unc-TTP are more informative than certainty examples. Furthermore, the Unc-TTP-guided uncertainty-based active example selection strategy outperforms existing methods, highlighting its effectiveness in classifying LLM uncertainty and enhancing in-context learning. This work not only underscores the potential of inconsistency-based uncertainty classification for both open- and closed-source LLMs but also presents a practical approach for leveraging uncertainty to improve LLM performance in real-world tasks.
♻ ☆ ObfuscaTune: Obfuscated Offsite Fine-tuning and Inference of Proprietary LLMs on Private Datasets AAAI 2025
This work addresses the timely yet underexplored problem of performing inference and finetuning of a proprietary LLM owned by a model provider entity on the confidential/private data of another data owner entity, in a way that ensures the confidentiality of both the model and the data. Hereby, the finetuning is conducted offsite, i.e., on the computation infrastructure of a third-party cloud provider. We tackle this problem by proposing ObfuscaTune, a novel, efficient and fully utility-preserving approach that combines a simple yet effective obfuscation technique with an efficient usage of confidential computing (only 5% of the model parameters are placed on TEE). We empirically demonstrate the effectiveness of ObfuscaTune by validating it on GPT-2 models with different sizes on four NLP benchmark datasets. Finally, we compare to a na\"ive version of our approach to highlight the necessity of using random matrices with low condition numbers in our approach to reduce errors induced by the obfuscation.
comment: Accepted at AAAI 2025 (PPAI Workshop)
♻ ☆ EEG-based AI-BCI Wheelchair Advancement: A Brain-Computer Interfacing Wheelchair System Using Deep Learning Approach
This study offers a revolutionary strategy to developing wheelchairs based on the Brain-Computer Interface (BCI) that incorporates Artificial Intelligence (AI) using a The device uses electroencephalogram (EEG) data to mimic wheelchair navigation. Five different models were trained on a pre-filtered dataset that was divided into fixed-length windows using a sliding window technique. Each window contained statistical measurements, FFT coefficients for different frequency bands, and a label identifying the activity carried out during that window that was taken from an open-source Kaggle repository. The XGBoost model outperformed the other models, CatBoost, GRU, SVC, and XGBoost, with an accuracy of 60%. The CatBoost model with a major difference between training and testing accuracy shows overfitting, and similarly, the best-performing model, with SVC, was implemented in a tkinter GUI. The wheelchair movement could be simulated in various directions, and a Raspberry Pi-powered wheelchair system for brain-computer interface is proposed here.
Robotics 17
☆ Learning Implicit Social Navigation Behavior using Deep Inverse Reinforcement Learning
This paper reports on learning a reward map for social navigation in dynamic environments where the robot can reason about its path at any time, given agents' trajectories and scene geometry. Humans navigating in dense and dynamic indoor environments often work with several implied social rules. A rule-based approach fails to model all possible interactions between humans, robots, and scenes. We propose a novel Smooth Maximum Entropy Deep Inverse Reinforcement Learning (S-MEDIRL) algorithm that can extrapolate beyond expert demos to better encode scene navigability from few-shot demonstrations. The agent learns to predict the cost maps reasoning on trajectory data and scene geometry. The agent samples a trajectory that is then executed using a local crowd navigation controller. We present results in a photo-realistic simulation environment, with a robot and a human navigating a narrow crossing scenario. The robot implicitly learns to exhibit social behaviors such as yielding to oncoming traffic and avoiding deadlocks. We compare the proposed approach to the popular model-based crowd navigation algorithm ORCA and a rule-based agent that exhibits yielding.
comment: 8 pages, Submitted to IEEE Robotics and Automation Letters (RAL)
☆ Shake-VLA: Vision-Language-Action Model-Based System for Bimanual Robotic Manipulations and Liquid Mixing
This paper introduces Shake-VLA, a Vision-Language-Action (VLA) model-based system designed to enable bimanual robotic manipulation for automated cocktail preparation. The system integrates a vision module for detecting ingredient bottles and reading labels, a speech-to-text module for interpreting user commands, and a language model to generate task-specific robotic instructions. Force Torque (FT) sensors are employed to precisely measure the quantity of liquid poured, ensuring accuracy in ingredient proportions during the mixing process. The system architecture includes a Retrieval-Augmented Generation (RAG) module for accessing and adapting recipes, an anomaly detection mechanism to address ingredient availability issues, and bimanual robotic arms for dexterous manipulation. Experimental evaluations demonstrated a high success rate across system components, with the speech-to-text module achieving a 93% success rate in noisy environments, the vision module attaining a 91% success rate in object and label detection in cluttered environment, the anomaly module successfully identified 95% of discrepancies between detected ingredients and recipe requirements, and the system achieved an overall success rate of 100% in preparing cocktails, from recipe formulation to action generation.
comment: Accepted to IEEE/ACM HRI 2025
☆ From Simulation to Field: Learning Terrain Traversability for Real-World Deployment
The challenge of traversability estimation is a crucial aspect of autonomous navigation in unstructured outdoor environments such as forests. It involves determining whether certain areas are passable or risky for robots, taking into account factors like terrain irregularities, slopes, and potential obstacles. The majority of current methods for traversability estimation operate on the assumption of an offline computation, overlooking the significant influence of the robot's heading direction on accurate traversability estimates. In this work, we introduce a deep neural network that uses detailed geometric environmental data together with the robot's recent movement characteristics. This fusion enables the generation of robot direction awareness and continuous traversability estimates, essential for enhancing robot autonomy in challenging terrains like dense forests. The efficacy and significance of our approach are underscored by experiments conducted on both simulated and real robotic platforms in various environments, yielding quantitatively superior performance results compared to existing methods. Moreover, we demonstrate that our method, trained exclusively in a high-fidelity simulated setting, can accurately predict traversability in real-world applications without any real data collection. Our experiments showcase the advantages of our method for optimizing path-planning and exploration tasks within difficult outdoor environments, underscoring its practicality for effective, real-world robotic navigation. In the spirit of collaborative advancement, we have made the code implementation available to the public.
comment: 38 pages
☆ ActiveGAMER: Active GAussian Mapping through Efficient Rendering
We introduce ActiveGAMER, an active mapping system that utilizes 3D Gaussian Splatting (3DGS) to achieve high-quality, real-time scene mapping and exploration. Unlike traditional NeRF-based methods, which are computationally demanding and restrict active mapping performance, our approach leverages the efficient rendering capabilities of 3DGS, allowing effective and efficient exploration in complex environments. The core of our system is a rendering-based information gain module that dynamically identifies the most informative viewpoints for next-best-view planning, enhancing both geometric and photometric reconstruction accuracy. ActiveGAMER also integrates a carefully balanced framework, combining coarse-to-fine exploration, post-refinement, and a global-local keyframe selection strategy to maximize reconstruction completeness and fidelity. Our system autonomously explores and reconstructs environments with state-of-the-art geometric and photometric accuracy and completeness, significantly surpassing existing approaches in both aspects. Extensive evaluations on benchmark datasets such as Replica and MP3D highlight ActiveGAMER's effectiveness in active mapping tasks.
☆ Toward a Universal Concept of Artificial Personality: Implementing Robotic Personality in a Kinova Arm
The fundamental role of personality in shaping interactions is increasingly being exploited in robotics. A carefully designed robotic personality has been shown to improve several key aspects of Human-Robot Interaction (HRI). However, the fragmentation and rigidity of existing approaches reveal even greater challenges when applied to non-humanoid robots. On one hand, the state of the art is very dispersed; on the other hand, Industry 4.0 is moving towards a future where humans and industrial robots are going to coexist. In this context, the proper design of a robotic personality can lead to more successful interactions. This research takes a first step in that direction by integrating a comprehensive cognitive architecture built upon the definition of robotic personality - validated on humanoid robots - into a robotic Kinova Jaco2 arm. The robot personality is defined through the cognitive architecture as a vector in the three-dimensional space encompassing Conscientiousness, Extroversion, and Agreeableness, affecting how actions are executed, the action selection process, and the internal reaction to environmental stimuli. Our main objective is to determine whether users perceive distinct personalities in the robot, regardless of its shape, and to understand the role language plays in shaping these perceptions. To achieve this, we conducted a user study comprising 144 sessions of a collaborative game between a Kinova Jaco2 arm and participants, where the robot's behavior was influenced by its assigned personality. Furthermore, we compared two conditions: in the first, the robot communicated solely through gestures and action choices, while in the second, it also utilized verbal interaction.
☆ Accelerating Discovery in Natural Science Laboratories with AI and Robotics: Perspectives and Challenges from the 2024 IEEE ICRA Workshop, Yokohama, Japan
Science laboratory automation enables accelerated discovery in life sciences and materials. However, it requires interdisciplinary collaboration to address challenges such as robust and flexible autonomy, reproducibility, throughput, standardization, the role of human scientists, and ethics. This article highlights these issues, reflecting perspectives from leading experts in laboratory automation across different disciplines of the natural sciences.
☆ Soft Vision-Based Tactile-Enabled SixthFinger: Advancing Daily Objects Manipulation for Stroke Survivors
The presence of post-stroke grasping deficiencies highlights the critical need for the development and implementation of advanced compensatory strategies. This paper introduces a novel system to aid chronic stroke survivors through the development of a soft, vision-based, tactile-enabled extra robotic finger. By incorporating vision-based tactile sensing, the system autonomously adjusts grip force in response to slippage detection. This synergy not only ensures mechanical stability but also enriches tactile feedback, mimicking the dynamics of human-object interactions. At the core of our approach is a transformer-based framework trained on a comprehensive tactile dataset encompassing objects with a wide range of morphological properties, including variations in shape, size, weight, texture, and hardness. Furthermore, we validated the system's robustness in real-world applications, where it successfully manipulated various everyday objects. The promising results highlight the potential of this approach to improve the quality of life for stroke survivors.
comment: Robosoft 2025 conference
☆ Cost-Effective Robotic Handwriting System with AI Integration
This paper introduces a cost-effective robotic handwriting system designed to replicate human-like handwriting with high precision. Combining a Raspberry Pi Pico microcontroller, 3D-printed components, and a machine learning-based handwriting generation model implemented via TensorFlow.js, the system converts user-supplied text into realistic stroke trajectories. By leveraging lightweight 3D-printed materials and efficient mechanical designs, the system achieves a total hardware cost of approximately \$56, significantly undercutting commercial alternatives. Experimental evaluations demonstrate handwriting precision within $\pm$0.3 millimeters and a writing speed of approximately 200 mm/min, positioning the system as a viable solution for educational, research, and assistive applications. This study seeks to lower the barriers to personalized handwriting technologies, making them accessible to a broader audience.
comment: This is an updated version of a paper originally presented at the 2024 IEEE Long Island Systems, Applications and Technology Conference (LISAT)
☆ Hierarchical Sampling-based Planner with LTL Constraints and Text Prompting
This project introduces a hierarchical planner integrating Linear Temporal Logic (LTL) constraints with natural language prompting for robot motion planning. The framework decomposes maps into regions, generates directed graphs, and converts them into transition systems for high-level planning. Text instructions are translated into LTL formulas and converted to Deterministic Finite Automata (DFA) for sequential goal-reaching tasks while adhering to safety constraints. High-level plans, derived via Breadth-First Search (BFS), guide low-level planners like Exploring Random Trees (RRT) and Probabilistic Roadmaps (PRM) for obstacle-avoidant navigation along with LTL tasks. The approach demonstrates adaptability to various task complexities, though challenges such as graph construction overhead and suboptimal path generation remain. Future directions include extending to considering terrain conditions and incorporating higher-order dynamics.
comment: 8 pages, 17 figures
☆ Vid2Sim: Realistic and Interactive Simulation from Video for Urban Navigation
Sim-to-real gap has long posed a significant challenge for robot learning in simulation, preventing the deployment of learned models in the real world. Previous work has primarily focused on domain randomization and system identification to mitigate this gap. However, these methods are often limited by the inherent constraints of the simulation and graphics engines. In this work, we propose Vid2Sim, a novel framework that effectively bridges the sim2real gap through a scalable and cost-efficient real2sim pipeline for neural 3D scene reconstruction and simulation. Given a monocular video as input, Vid2Sim can generate photorealistic and physically interactable 3D simulation environments to enable the reinforcement learning of visual navigation agents in complex urban environments. Extensive experiments demonstrate that Vid2Sim significantly improves the performance of urban navigation in the digital twins and real world by 31.2% and 68.3% in success rate compared with agents trained with prior simulation methods.
comment: Project page: https://metadriverse.github.io/vid2sim/
☆ Application of Vision-Language Model to Pedestrians Behavior and Scene Understanding in Autonomous Driving
Autonomous driving (AD) has experienced significant improvements in recent years and achieved promising 3D detection, classification, and localization results. However, many challenges remain, e.g. semantic understanding of pedestrians' behaviors, and downstream handling for pedestrian interactions. Recent studies in applications of Large Language Models (LLM) and Vision-Language Models (VLM) have achieved promising results in scene understanding and high-level maneuver planning in diverse traffic scenarios. However, deploying the billion-parameter LLMs to vehicles requires significant computation and memory resources. In this paper, we analyzed effective knowledge distillation of semantic labels to smaller Vision networks, which can be used for the semantic representation of complex scenes for downstream decision-making for planning and control.
♻ ☆ High-Sensitivity Vision-Based Tactile Sensing Enhanced by Microstructures and Lightweight CNN
Tactile sensing is critical in advanced interactive systems by emulating the human sense of touch to detect stimuli. Vision-based tactile sensors (VBTSs) are promising for their ability to provide rich information, robustness, adaptability, low cost, and multimodal capabilities. However, current technologies still have limitations in sensitivity, spatial resolution, and the high computational demands of deep learning-based image processing. This paper presents a comprehensive approach combining a novel sensor structure with micromachined structures and an efficient image processing method, and demonstrates that carefully engineered microstructures within the sensor hardware can significantly enhance sensitivity while reducing computational load. Unlike traditional designs with tracking markers, our sensor incorporates an interface surface with micromachined trenches, as an example of microstructures, which modulate light transmission and amplify the variation in response to applied force. By capturing variations in brightness, wire width, and cross pattern locations with a camera, the sensor accurately infers the contact location, the magnitude of displacement and applied force with a lightweight convolutional neural network (CNN). Theoretical and experimental results demonstrated that the microstructures significantly enhance sensitivity by amplifying the visual effects of shape distortion. The sensor system effectively detected forces below 10 mN, and achieved a millimetre-level single-point spatial resolution. Using a model with only one convolutional layer, a mean absolute error (MAE) below 0.05 mm have been achieved. Its soft sensor body ensures compatibility with soft robots and wearable electronics, while its immunity to electrical crosstalk and interference guarantees reliability in complex human-machine environments.
comment: 27 pages, 13 figures, 2 tables; rearranged figures; corrected typos
♻ ☆ A Survey on Reinforcement Learning Applications in SLAM
The emergence of mobile robotics, particularly in the automotive industry, introduces a promising era of enriched user experiences and adept handling of complex navigation challenges. The realization of these advancements necessitates a focused technological effort and the successful execution of numerous intricate tasks, particularly in the critical domain of Simultaneous Localization and Mapping (SLAM). Various artificial intelligence (AI) methodologies, such as deep learning and reinforcement learning, present viable solutions to address the challenges in SLAM. This study specifically explores the application of reinforcement learning in the context of SLAM. By enabling the agent (the robot) to iteratively interact with and receive feedback from its environment, reinforcement learning facilitates the acquisition of navigation and mapping skills, thereby enhancing the robot's decision-making capabilities. This approach offers several advantages, including improved navigation proficiency, increased resilience, reduced dependence on sensor precision, and refinement of the decision-making process. The findings of this study, which provide an overview of reinforcement learning's utilization in SLAM, reveal significant advancements in the field. The investigation also highlights the evolution and innovative integration of these techniques.
♻ ☆ Beyond Sight: Finetuning Generalist Robot Policies with Heterogeneous Sensors via Language Grounding
Interacting with the world is a multi-sensory experience: achieving effective general-purpose interaction requires making use of all available modalities -- including vision, touch, and audio -- to fill in gaps from partial observation. For example, when vision is occluded reaching into a bag, a robot should rely on its senses of touch and sound. However, state-of-the-art generalist robot policies are typically trained on large datasets to predict robot actions solely from visual and proprioceptive observations. In this work, we propose FuSe, a novel approach that enables finetuning visuomotor generalist policies on heterogeneous sensor modalities for which large datasets are not readily available by leveraging natural language as a common cross-modal grounding. We combine a multimodal contrastive loss with a sensory-grounded language generation loss to encode high-level semantics. In the context of robot manipulation, we show that FuSe enables performing challenging tasks that require reasoning jointly over modalities such as vision, touch, and sound in a zero-shot setting, such as multimodal prompting, compositional cross-modal prompting, and descriptions of objects it interacts with. We show that the same recipe is applicable to widely different generalist policies, including both diffusion-based generalist policies and large vision-language-action (VLA) models. Extensive experiments in the real world show that FuSeis able to increase success rates by over 20% compared to all considered baselines.
♻ ☆ An Accurate and Real-time Relative Pose Estimation from Triple Point-line Images by Decoupling Rotation and Translation
Line features are valid complements for point features in man-made environments. 3D-2D constraints provided by line features have been widely used in Visual Odometry (VO) and Structure-from-Motion (SfM) systems. However, how to accurately solve three-view relative motion only with 2D observations of points and lines in real time has not been fully explored. In this paper, we propose a novel three-view pose solver based on rotation-translation decoupled estimation. First, a high-precision rotation estimation method based on normal vector coplanarity constraints that consider the uncertainty of observations is proposed, which can be solved by Levenberg-Marquardt (LM) algorithm efficiently. Second, a robust linear translation constraint that minimizes the degree of the rotation components and feature observation components in equations is elaborately designed for estimating translations accurately. Experiments on synthetic data and real-world data show that the proposed approach improves both rotation and translation accuracy compared to the classical trifocal-tensor-based method and the state-of-the-art two-view algorithm in outdoor and indoor environments.
♻ ☆ USV-AUV Collaboration Framework for Underwater Tasks under Extreme Sea Conditions
Autonomous underwater vehicles (AUVs) are valuable for ocean exploration due to their flexibility and ability to carry communication and detection units. Nevertheless, AUVs alone often face challenges in harsh and extreme sea conditions. This study introduces a unmanned surface vehicle (USV)-AUV collaboration framework, which includes high-precision multi-AUV positioning using USV path planning via Fisher information matrix optimization and reinforcement learning for multi-AUV cooperative tasks. Applied to a multi-AUV underwater data collection task scenario, extensive simulations validate the framework's feasibility and superior performance, highlighting exceptional coordination and robustness under extreme sea conditions. To accelerate relevant research in this field, we have made the simulation code (demo version) available as open-source.
♻ ☆ Speedup Techniques for Switchable Temporal Plan Graph Optimization AAAI 2025
Multi-Agent Path Finding (MAPF) focuses on planning collision-free paths for multiple agents. However, during the execution of a MAPF plan, agents may encounter unexpected delays, which can lead to inefficiencies, deadlocks, or even collisions. To address these issues, the Switchable Temporal Plan Graph provides a framework for finding an acyclic Temporal Plan Graph with the minimum execution cost under delays, ensuring deadlock- and collision-free execution. Unfortunately, existing optimal algorithms, such as Mixed Integer Linear Programming and Graph-Based Switchable Edge Search (GSES), are often too slow for practical use. This paper introduces Improved GSES, which significantly accelerates GSES through four speedup techniques: stronger admissible heuristics, edge grouping, prioritized branching, and incremental implementation. Experiments conducted on four different map types with varying numbers of agents demonstrate that Improved GSES consistently achieves over twice the success rate of GSES and delivers up to a 30-fold speedup on instances where both methods successfully find solutions.
comment: Accepted by AAAI 2025. This version contains the appendix
Multimedia 2
♻ ☆ Exploring Transferability of Multimodal Adversarial Samples for Vision-Language Pre-training Models with Contrastive Learning
The integration of visual and textual data in Vision-Language Pre-training (VLP) models is crucial for enhancing vision-language understanding. However, the adversarial robustness of these models, especially in the alignment of image-text features, has not yet been sufficiently explored. In this paper, we introduce a novel gradient-based multimodal adversarial attack method, underpinned by contrastive learning, to improve the transferability of multimodal adversarial samples in VLP models. This method concurrently generates adversarial texts and images within imperceptive perturbation, employing both image-text and intra-modal contrastive loss. We evaluate the effectiveness of our approach on image-text retrieval and visual entailment tasks, using publicly available datasets in a black-box setting. Extensive experiments indicate a significant advancement over existing single-modal transfer-based adversarial attack methods and current multimodal adversarial attack approaches.
♻ ☆ CAMSIC: Content-aware Masked Image Modeling Transformer for Stereo Image Compression AAAI 2025
Existing learning-based stereo image codec adopt sophisticated transformation with simple entropy models derived from single image codecs to encode latent representations. However, those entropy models struggle to effectively capture the spatial-disparity characteristics inherent in stereo images, which leads to suboptimal rate-distortion results. In this paper, we propose a stereo image compression framework, named CAMSIC. CAMSIC independently transforms each image to latent representation and employs a powerful decoder-free Transformer entropy model to capture both spatial and disparity dependencies, by introducing a novel content-aware masked image modeling (MIM) technique. Our content-aware MIM facilitates efficient bidirectional interaction between prior information and estimated tokens, which naturally obviates the need for an extra Transformer decoder. Experiments show that our stereo image codec achieves state-of-the-art rate-distortion performance on two stereo image datasets Cityscapes and InStereo2K with fast encoding and decoding speed. Code is available at https://github.com/Xinjie-Q/CAMSIC.
comment: Accepted by AAAI 2025